McGuire, Units 1 and 2, Response to Request for Additional ... · tags/labels) that transport to the ECCS sump are addressed in Enclosure 2, Section 3(d). The following assumptions

BRUCE H HAMILTONDuke Vice President

AhEnergye McGuire Nuclear Station

Duke Energy CorporationMG01 VP / 12700 Hagers Ferry-RoadHuntersville, NC 28078

704-875-5333

704-875-4809 faxbhhamilton@duke-energy. com

February 28, 2008

U. S. Nuclear Regulatory CommissionWashington, DC 20555-0001ATTENTION: Document Control Desk

SUBJECT: Duke Energy CorporationMcGuire Nuclear Station, Units 1 and 2Docket Nos. 50-369 and 50-370NRC Generic Letter 2004-02, "Potential Impact of Debris Blockageon Emergency Recirculation During Design Basis Accidents atPressurized-Water Reactors"

On September 13, 2004, the Nuclear Regulatory Commission (NRC) issuedGeneric Letter (GL) 2004-02. The GL requested that all pressurized-waterreactor (PWR) licensees (1) evaluate the adequacy of the emergency sumprecirculation function with respect to potentially adverse effects associated withpost-accident debris, and (2) implement any plant modifications determined to benecessary.

By letter dated March 1, 2005, as supplemented by letter dated September 1,2005, Duke Power Company LLC d.b.a. Duke Energy Carolinas, LLC (Duke)provided responses to GL 2004-02. By letter dated February 9, 2006, the NRCdetermined that additional information was necessary in order for the Staff tocomplete their review of McGuire's information. McGuire's responses to theserequests for additional information are contained in Enclosure 1.

On November 30, 2007, the NRC issued a letter to the Nuclear Energy Instituteauthorizing all PWR licensees up to two months beyond December 31, 2007 (i.e.,to February 29, 2008), to provide the supplemental responses to the NRC.

Additionally, by letter dated November 21, 2007, the NRC staff issued a revised"Content Guide for Generic Letter 2004-02 Supplemental Responses" for the useby PWR licensees in developing their GL 2004-02 responses. McGuire'ssupplemental responses are contained in Enclosure 2.

As stated by Duke's letter of November 6, 2007, as amended by letter datedDecember 13, 2007, any additional or revised information resulting from the

www. duke-energy. com

US Nuclear Regulatory CommissionFebruary 28, 2008Page 2

Integrated Prototype (chemical effects) Testing will be provided as an amendedresponse to GIL 2004-02 by April 30, 2008. This extension was approved by thestaff in a letter dated December 28, 2007. Additionally, as requested within thisapproval letter, Duke will provide additional information related to the NRC staff-requested evaluation of WCAP-1 6406, Revision 1 dated August 2007.

Duke understands that the NRC staff will consider this set of additionalinformation and will issue a letter to Duke Energy assessing the overall adequacyof the McGuire Station's GIL 2004-02 corrective actions.

If any questions arise or additional information is needed, please contact K. L.Ashe at (704) 875-4535.

Very truly yours,

Bruce H. Hamilton

Enclosures


Bruce H. Hamilton affirms that he is the person who subscribed his name to theforegoing statement, and that all the matters and facts set forth herein are trueand correct to the best of his knowledge.

Bruce H. Hamilton, Vice President, McGuire Nuclear Station

Subscribed and sworn to me: 9 S• )Date

oýoý aa'ý.4' L' -- Y•r, ,Notary Public

q-1 -2 0/ a-_.,My commission expires:

Date

SEAL


xc: w/enclosures

V. M McCree, Acting Region II Administrator(Acting) U.S. Nuclear Regulatory CommissionSam Nunn Atlanta Federal Center, 23 T8561 Forsyth St., SWAtlanta, GA 30303-8931

J. F. Stang, Jr., Senior Project Manager (MNS)U'. S. Nuclear Regulatory Commission11555 Rockville PikeMail Stop 0-8 G9ARockville, MD 20852-2738

J. B. BradyNRC Senior Resident InspectorMcGuire Nuclear Station

B. 0. Hall, Senior ChiefDivision of Radiation Section1645 Mail Service CenterRaleigh, NC 27699-1645

Enclosure 1

Response to Request for Additional Information

Enclosure 1Responses to Staff Request for Additional Information Identified on February 9, 2006

McGuire Nuclear Station Units 1 and 2Generic Letter 2004-02

Request for Additional Information IIdentify the name and bounding quantity of each insulation material generatedby a large-break loss-of-coolant accident (LBLOCA). Include the amount ofthese materials transported to the containment pool. State any assumptionsused to provide this response.

McGuire Response:General Note:This RAI response describes the initial, unrefined quantities of insulation debris usedin the baseline evaluation for sizing the McGuire ECCS sump strainer. Refinedquantities of fibrous insulation debris used in design validation will be addressed asstated in response to RAI 12 of Enclosure 1.

Table 1-1 is a summary of the McGuire Nuclear Station insulation debris generatedand transported to the ECCS sump by a large break loss of coolant accident. Thelimiting case is the "B" loop hot leg break.

Table 1-1Insulation Debris Values

Debris Type Break Debris Quantity Debris Quantity AtZone of Generated Transport Sump

Influence Fraction(ZOI) (DTF)

Insulation(Nukon® and Thermal-Wrap®)

Low Density Fiberglass (LDFG)Fines 17D 272.7 ftW 100% 272.7 ftWSmall Pieces (<6" on a Side) 17D 891.9 ftW 21% 187.3 ft"Large Pieces (>6" on a Side) 17D 444.9 ft2 10% 44.5 ft2

Intact Blankets 17D 476.4 ftW 0% 0 ft2

Reflective Metal Insulation (RMI)Small Pieces (<4") 28.6D 22,246 ftz 0% 0 ft2

Large Pieces (>4") 28.6D 9,087 ft7 0% 0 ft2

I A I. I

Note: Only insulation debris is addressed in this RAI response. The quantities offailed coatings that transport to the ECCS sump are addressed separately inEnclosure 2, Section 3(h). Latent debris quantities (fiber fines, dust/dirt fines,tags/labels) that transport to the ECCS sump are addressed in Enclosure 2, Section3(d).

The following assumptions have been made regarding debris qeneration:

Page 1 of 38



1. It was assumed that buckles, straps, and wires securing insulation would nottransport, and therefore can be excluded from the debris source term. Thesematerials are metal based, and would readily sink to the floor. Also, the volumeof these materials is negligible compared to the insulation volume.

2. It was assumed that all jacketed insulation outside of the ZOI would not undergoany erosion by either break or spray flows, i.e., no insulation debris would begenerated outside of the postulated ZOls.

3. Since RMI is not a significant contributor to head loss compared to a fibrousdebris bed, an exact quantification of RMI for the break was not required. Largeequipment (reactor coolant pumps and pressurizer) and large bore piping, e.g.,RCS, Main Steam, Feedwater, and Auxiliary Feedwater, were considered in thetabulation of RMI. Even if the foil area of the RMI is significantly changed, itscontribution to the total sump strainer head loss is negligible.

4. Thermal-Wrap® is a low density fiberglass insulation similar to Nukon®. Thematerial characteristics, as well as destruction pressure and associated ZOI, areassumed equal to those defined for Nukon®.

The following assumptions have been made regarding debris transport:

5. NUKON® and Thermal-Wrap® are identical for transport purposes. This is areasonable assumption since both products are low density fiberglass withsimilar material properties.

6. It was assumed that small pieces of fiberglass (smaller than 6") can be treatedas 1" clumps, and large pieces can be treated as 6" pieces for transportpurposes. This is a conservative assumption since smaller pieces of fiberglasstransport more readily than larger pieces.

7. It was assumed that 1/4" - 4" pieces of RMI can be treated as 1/2" pieces, and4"- 6" pieces can be treated as 2" pieces for transport purposes. This is aconservative assumption since smaller pieces of RMI transport more readilythan larger pieces.

8. It was assumed that RMI would not break down into smaller pieces following theinitial generation. This is a reasonable assumption since RMI is a metallicinsulation that would not be subject to erosion by the flow of water.

9. It was assumed that the settling velocity of fine debris (insulation, dirt/dust, andpaint particulate) can be calculated using Stokes' law. This is a reasonable

Page 2 of 38



assumption since the particulate is generally spherical and would settle slowly(within the applicability of Stokes' law).

10. Based on fibrous debris testing, it was assumed that the fiberglass debris wouldnot float in the containment pool. Test data has shown that fiberglass insulationsinks more readily in hotter water. Therefore, given the initial high temperatureof the containment pool at McGuire (190°F), this is a reasonable assumption.

11. It was conservatively assumed that all of the debris generated by the postulatedletdown line break would be transported to the sump (the letdown line break isnot limiting even with this conservative assumption).

12. It was assumed that the fine debris was uniformly distributed in the pool at thebeginning of recirculation.

13. With the exception of the debris blown through the crane wall penetrations, itwas assumed that small and large piece debris would be uniformly distributedinside the crane wall.

14. It was assumed that the recirculation transport fractions determined for the Loop"B" break can be applied to the other breaks inside the crane wall. This isconservative, since the Loop "B" is closest to the sump.

15. Water falling from the reactor coolant system was conservatively assumed to doso without encountering any structures before reaching the containment pool.

16. It was assumed that a fraction of the fine debris as well as the small and largepiece debris would be carried through the crane wall penetrations to the pipechase in proportion to the blowdown flow split to the pipe chase.

17. It was conservatively assumed that all debris blown into the ice condenser wouldbe subsequently washed back down with the melting ice flow.

18. For the Computational Fluid Dynamics (CFD) model, it was assumed thatpotential upstream blockage points.(e.g. drains, grating, etc.) would not inhibitthe flow of water through these areas.

Page 3 of 38



Request for Additional Information 2Identify the amounts (i.e., surface area) of the following materials that are:

a. submerged in the containment pool following a loss-of-coolant accident(LOCA),

b. in the containment spray zone following a LOCA:

- aluminum

- zinc (from galvanized steel and from inorganic zinc coatings)

- copper- carbon steel not coated

- uncoated concrete

Compare the amounts of these materials in the submerged and spray zones atyour plant relative to the scaled amounts of these materials used in theNuclear Regulatory Commission (NRC) nuclear industry jointly-sponsoredIntegrated Chemical Effects Tests (ICET) (e.g., 5x the amount of uncoatedcarbon steel assumed for the ICETs).

McGuire Response:General note:The published results of ICET Test #5 were used in developing the input parametersof the Integrated Prototype Test (IPT) described in the response to RAI #11 ofEnclosure 1. ICET Test #5 is not used directly to assess the chemical effects on theMcGuire strainer head loss. The Duke IPT is a separate, comprehensive chemicaleffects test that emulates a portion of the ICET Test #5 battery, using bounding andrepresentative chemical and debris input parameters that vary as a function of theECCS mission time (to simulate the effects of spray) and are more closely coupledto the predicted McGuire post-LOCA environment.

The following is an assessment of the amount of the materials identified by RAI #2located in the submerged zone (i.e., in the containment sump pool) in the post-LOCA environment at McGuire:

AluminumA bounding estimate of the amount of aluminum expected to be submerged inthe containment sump pool following a LOCA is 530 square feet.

* Zinc (from galvanized steel and from inorganic zinc coatings)In an ice condenser containment such as McGuire, the galvanized steelcomponents (e.g., baskets and structural steel) located within the boundary ofthe ice condenser itself are outside both the submergence and thecontainment spray zones. The total remaining estimated zinc inventory, in theform of metallic coatings, zinc based coatings, and electrical components and

Page 4 of 38



equipment, is conservatively estimated at 543,700 square feet. A boundingestimate of the amount of this remaining inventory expected to be submergedin the post-LOCA containment pool is 142,800 square feet.

CopperThe copper inventory inside containment is not specifically tracked atMcGuire.

Uncoated carbon steelThe uncoated carbon steel surface area inside the McGuire containments isnot specifically tracked.

Uncoated concreteUncoated concrete surface areas inside the containments at McGuire are notspecifically tracked. In general, all concrete surfaces inside containment arecoated; however, there are some inaccessible areas that cannot be confirmedto.have coatings (e.g., clearance between concrete expansion joints, thecavity between the reactor vessel and bio-shield wall). Since the majority ofthe ECCS sump is coated, only a minor fraction of uncoated concrete surfacearea would be submerged in the post-LOCA containment sump pool.

The following is an assessment of the amount of the materials identified by RAI #2located in the spray zone (i.e., not submerged in the containment sump pool, butexposed to containment spray flow) in the post-LOCA environment at McGuire:

AluminumA bounding estimate of the amount of aluminum identified as in the sprayzone following a LOCA is 740 square feet.

Zinc (from galvanized steel and from inorganic zinc coatings)There is estimated to be 100,900 square feet of zinc inventory in the sprayzone following a LOCA (i.e., miscellaneous galvanized steel and electricalsupport components). There is 300,000 square feet of top-coated zincprimers (qualified coatings) that are not considered a contributor to post-LOCA containment pool chemistry. Qualified coatings are only a containmentpool particulate debris concern in the limiting LBLOCA coatings ZOI.

* CopperThe copper inventory inside containment is not specifically tracked atMcGuire.

* Uncoated carbon steelThe uncoated carbon steel surface area inside the McGuire containments isnot specifically tracked.

Page 5 of 38



Uncoated concreteUncoated concrete surface areas inside the containments at McGuire are notspecifically tracked. In general, all concrete surfaces inside containment arecoated, however there are some inaccessible areas that cannot be confirmedto have coatings (e.g., clearance between concrete expansion joints, thecavity between the reactor vessel and bio-shield wall). While some fraction ofthis uncoated concrete will be above the ECCS sump and in the containmentspray zone, the relative inaccessibility of this uncoated concrete surface areaminimizes the effect of containment spray exposure.

A comparison of the expected amounts of these materials in the submergedcontainment pool zones relative to the scaled amounts of these materials used inICET Test #5 (for ice condenser plants) is summarized in Table 2-1.

A comparison of the expected amounts of these materials exposed to containmentspray (i.e., unsubmerged) does not yield additional McGuire/ICET Test #5comparison information beyond that depicted in Table 2-1.

The expected McGuire post-LOCA containment sump pool chemistry (boronconcentration, buffering agent concentration, and pH) is compared to the ICET Test#5 conditions in the response to RAI #6 of Enclosure 1.

Page 6 of 38



Table 2-1McGuire Values vs. ICET Test #5 Values

MNS Values Values used in ICETTest #5

Amount Amount Ratio* Test SubmergedParameter (total Submerged Ratio Material

submerged + (from boundingexposed to estimates)**

spraAi**

Aluminum ft 2 600 f 2 0.03 3.51300 (46%) ft2/ft 3 ft2/ft3

Zinc in 150,000 ft2 7.4 8.0Galvanized 250,000 ft2 (60%) ft2/ft3 ft2/ft3

SteelInorganic Zinc

Primer 4.6Coatings'(non- See Note 1 See Note 1 ft2/ft3

top coated)Copper

(including Cu- Not Tracked Not Tracked 6.0Ni alloys) (see Note 2) (see Note 2) ft2/ft3 25%

Carbon SteelNot Tracked Not Tracked 0.15(see Note 3) (see Note 3) ft2/ft 3

Concrete(surface, Not Tracked Not Tracked 0.045

uncoated) (see Note 4) (see Note 4) ft2/ft 3 34%

* McGuire minimum ECCS sump pool volume used (20,234 cubic feet) to maximize the ratio

for ICET Test #5 comparison.** These values are rounded up from the values in the response to RAI #2 for comparisonpurposes to the ICET Test #5 scaled amounts.

Note 1:McGuire does not utilize inorganic zinc coatings, therefore the material is not considered in thiscbmparison.

Page 7 of 38



Note 2:As demonstrated during ICET Test #5 and as identified in WCAP-16530-NP, copper and Cu-Ni alloysare resistant to corrosion under expected post-accident conditions and appear in only trace amountsin the predicted sump pool chemistry.

Note 3:Carbon steel is a metal alloy primarily composed of iron. As demonstrated during ICET Test #5, ironappears in only trace amounts in the predicted sump pool chemistry1 °. Further, it was identified inWCAP-1 6530-NP that the release rates for iron were small and subsequently ignored in chemicaleffects precipitation modeling.

Note 4:Uncoated concrete occurs in limited amounts in containment. As demonstrated during ICET Test #5,concrete is primarily a particulate debris concern in the McGuire post-LOCA containment sump pool.Further, sensitivity tests performed under WCAP-1 6530-NP determined that the precipitation ofmaterials from concrete dissolution was negligible even with high exposed surface areas.

Based on the above assessments of the amount of the materials identified in RAI #2,the predicted post-LOCA conditions at McGuire compare reasonably andconservatively to the parameters and conditions tested in ICET Test #5. As notedearlier, the results from ICET Test #5 were used to help develop input parametersfor Duke's Integrated Prototype Test for chemical effects.

Request for Additional Information 3Identify the amount (surface area) and material (e.g., aluminum) for anyscaffolding stored in containment. Indicate the amount, if any, that would besubmerged in the containment pool following a LOCA. Clarify if scaffoldingmaterial was included in the response to Question 2.

McGuire Response:McGuire does not permanently store scaffold materials inside containment.

On occasion, scaffolds may be temporarily installed in containment during poweroperations to support specific maintenance activities. Installation of temporaryscaffolding is procedurally controlled. Aluminum inventory related to temporaryscaffolding installed inside containment is logged and compared to administrativelimits to ensure that the limits are not exceeded. Margin is included in the aluminumvalues provided in the response to RAI #2 of Enclosure 1 for temporary scaffoldingapplications.

Zinc (associated with galvanized coatings on certain scaffolding components) isprimarily a particulate concern in the post-LOCA ECCS sump pool and is aninsignificant contributor to post-LOCA sump chemistry. There is substantial margin inthe inventory limit for zinc inside containment; therefore, the zinc coatingsassociated with temporary scaffolding is not a specifically tracked element.

Page 8 of 38



Request for Additional Information 4Provide the type and amount of any metallic paints or non-stainless steelinsulation jacketing (not included in the response to Question 2) that would beeither submerged or subjected to containment spray.

McGuire Response:All insulation jacketing materials inside containment are stainless steel.

The only known metallic paint, other than zinc-based primers, is associated withtouch-up coatings on galvanized steel items. This coating application has had verylimited use and has been accounted for as an unqualified coating. Galvanized steelitems are accounted for in the response to RAI #2 of Enclosure 1.

Request for Additional Information 5Provide the expected containment pool pH during the emergency core coolingsystem (ECCS) recirculation mission time following a LOCA at the beginningof the fuel cycle and at the end of the fuel cycle. Identify any keyassumptions.

McGuire Response:In order to establish the containment sump pool pH at a particular time after theinitiation of a postulated LBLOCA event, the addition of water to the sump from thevarious Emergency Core Cooling System (ECCS) sources and the Ice Condensermust be established. The guidance from NUREG/CR-5950 was followed. Majorinputs and assumptions used in the McGuire Nuclear Station ECCS sump pHcalculation are provided in Table 5-1 below. LBLOCA is considered because of thelarge potential for debris generation.

An analysis was performed to document how the ECCS sump pH was affected byboth Beginning-of-Life (BOL) and End-of-Life (EOL) boron concentrations in thereactor coolant inventory. The EOL scenario, consisting of much lowerconcentrations of boron in the reactor coolant inventory, yields a higher sump pHprofile than the same analysis for the BOL assessment. Figure 5-1 shows the time-dependent sump pH profile for both BOL and EOL cases, normalized to 25 °C.

The McGuire ECCS sump pH analysis shows that by 2 hours after LBLOCA initiationall of the water from the RCS, Refueling Water Storage Tank (FWST), Cold LegAccumulators (CLAs), and the ice melt from the Ice Condenser will be in the sump.Continuing nitric acid and hydrochloric acid production due to irradiation of thewater/air mixture and of the electrical cable insulation/jacket material insidecontainment will be a long term contributor to the pH value of the sump. In order toaddress the impact of this effect, a sensitivity analysis was performed and showed

Page 9 of 38



that when the time range is expanded beyond 2 hours, there is an insignificantchange in the ECCS sump pH.

Figure 5-1McGuire LBLOCA Expected Beginning and End of Life Sump pH Response

9.0 -

8.8 Normalized pH at 25 C - BOL

-...Normalized pH at 25 C - EOL

8.6

8.2

8.0

7.8

7.6

7.4

7.2

7.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Time (Minutes)

Page 10 of 38



Key Assumptions:Table 5-1

Key Assumptions For the Calculation of Post LOCAContainment Sump pH at McGuire Nuclear Station

Input Description Value

Reactor Coolant Inventory 11,114.4 ftW

Boron Content in Reactor Coolant (BOL) 1220.5 ppm

Boron Content in Reactor Coolant (EOL) 9.0 ppm

FWST Inventory 346,606 gal

Boron Content in FWST 2775 ppm

Cold Leg Accumulators Volume 3800 ftj

Boron Content in Cold Leg Accumulators 2675 ppm

Ice Condenser Mass 1,890,180 Ibm

Ice Condenser Boron Content 2065 ppm

Request for Additional Information 6For the ICET environment that is the most similar to your plant conditions,compare the expected containment pool conditions to the ICET conditions forthe following items: boron concentration, buffering agent concentration, andpH. Identify any other significant differences between the ICET environmentand the expected plant-specific environment.

McGuire Response:Post-LOCA containment sump pool chemistry is a function of Reactor CoolantSystem (RCS)/Emergency Core Cooling System (ECCS)/Ice Condenser inventorychemistry, and also a function of the chemical effects due to materials submerged inthe pool (and sprayed outside the pool) during the ECCS mission time. Theresponse to RAI #2 of Enclosure 1 addresses materials in the containment pool andthe response to RAI #5 of Enclosure 1 specifically addresses the pH of thecontainment pool. The multi-part pool chemistry comparison requested in this RAIinvokes the previous pH response, so that information is repeated here (in a differentform) for consistency. /

Page 11of 38



McGuire is an ice condenser plant using sodium tetraborate as the buffering agent.It has no calcium silicate insulation material. Therefore, the ICET environment that ismost similar to McGuire is ICET Test #5. A comparison of the expected McGuirecontainment pool conditions and ICET Test #5 conditions at fuel cycle Beginning-of-Life (BOL) and End-of-Life (EOL) is provided in Table 6-1:

Table 6-1McGuire Containment Pool Conditions vs. ICET Test #5 Conditions

McGuire BOL McGuire EOL ICET Test 5Boron 1703 to 2487 898 to 2372

Concentration >2013 after 1 minute >1592 after 1 minute 2400(ppm B) 2478 after 110 minutes 2370 after 88 minutes

6.55 to 8.37 7.04 to 8.62pH (normalized to <8 after 16 minutes <8 after 23 minutes

25C) 7.83-7.89 after 67 7.88-7.94 after 64 minutesminutes

Na concentrationfrom sodium 755 after 110 minutes 755 after 110 minutes 1200 to 1400tetraborate > 465 after 18 seconds > 465 after 18 seconds(ppm Na)

Based on the comparison above, ICET Test #5 parameters are representative of orbound the predicted McGuire post-LOCA environment parameters.

Page 12 of 38



Request for Additional Information 7For a LBLOCA, provide the time until ECCS external recirculation initiationand the associated pool temperature and pool volume. Provide estimated pooltemperature and pool volume 24 hours after a LBLOCA. Identify theassumptions used for these estimates.

McGuire Response:For LBLOCA the time until ECCS external recirculation initiation and the associatedpool temperatures and pool volumes are shown in the table below.

Table 7-1ECCS Parameters After LBLOCA

Analysis Time Sump Sump Volume (ft3)Assumptions Temperature (*F)Minimum 1579 seconds to 185 46,916Safeguards recirc. initiation(Note 1)

24 hours 145 57,819Maximum 806 seconds to 189 43,352Safeguards recirc. initiation(Note 2)

24 hours 114 59,972

Note 1: Minimum Safeguards conditions are characterized by minimum flow rates from one train ofECCS pumps, including one containment spray pump.

Note 2: Maximum Safeguards conditions are characterized by maximum flow rates from two trains ofECCS pumps, including two containment spray pumps.

Page 13 of 38



Request for Additional Information 8Discuss your overall strategy to evaluate potential chemical effects includingdemonstrating that, with chemical effects considered, there is sufficient netpositive suction head (NPSH) margin available during the ECCS mission time.Provide an estimated date with milestones for the completion of all chemicaleffects evaluations.

McGuire Response:Duke's overall strategy to evaluate chemical effects employs an Integrated PrototypeTest (IPT) that is designed to simulate the predicted comprehensive challenge to theMcGuire and Catawba ECCS Strainers in the post-LOCA containment pool. The IPTcombines the physical and chemical characteristics expected in the post-accidentenvironment, and then challenges a prototype strainer module in this environmentfor the full 30-day ECCS mission time. Overall characteristics of the IPT areidentified below:

" Full-scale strainer top-hat module

" Representative debris load (fiber)

" Bounding particulate load (coatings, dust/dirt)

* Bounding sump pool chemistry (pH, boron concentration)

* Representative sump temperature cool-down profile

* Bounding approach velocity

* Bounding dissolved aluminum concentration as a function of ECCS mission time

The input parameters for the IPT are representative of the McGuire and Catawbapost-LOCA sump conditions. Further details regarding these parameters can befound in the response to RAI #11 in Enclosure 1.

Available ECCS NPSH margin/head loss calculations have been documented for theclean strainer condition and the refined debris load condition in accordance with theapproved methodology identified in NEI 04-07 and the NRC SER. The refined debrisload head loss analysis does not yet incorporate chemical effects. Upon completionof the IPT documentation, a refined debris load head loss calculation will incorporateany added consequence of tested chemical effects and confirm the available ECCSNPSH margin. Duke expects to have this documentation and the McGuire Units 1and 2 head loss calculations completed by April 30, 2008.

Page 14 of 38



Request for Additional Information 9Identify, if applicable, any plans to remove certain materials from thecontainment building and/or to make a change from the existing chemicalsthat buffer containment pool pH following a LOCA.

McGuire Response:McGuire is an ice condenser plant, using sodium tetraborate in the ice condenser asa pH buffer. There are no plans to make any changes to the existing chemicals thatbuffer the McGuire containment pool pH following a LOCA.

Request for Additional Information 10If bench-top testing is being used to inform plant specific head loss testing,indicate how the bench-top test parameters (e.g., buffering agentconcentrations, pH, materials, etc.) compare to your plant conditions.Describe your plans for addressing uncertainties related to head loss fromchemical effects including, but not limited to, use of chemical surrogates,scaling of sample size and test durations. Discuss how it will be determinedthat allowances made for chemical effects are conservative.

McGuire Response:General Note:In order to address specific ECCS head loss/NPSH issues, Duke designed acomprehensive Integrated Prototype Test (IPT) to assess chemical effects on thepostulated post-LOCA debris bed at McGuire and Catawba. The design and set-upof the IPT is discussed in detail in the response to RAI #11 of Enclosure 1.

Bench-top tests (including laboratory tests and a 30-day Vertical Loop Test) werealso performed as part of an aluminum release rate testing program, in order toprovide insights for the Integrated Prototype Test parameters. The bench-top testingprogram that led to the development of the IPT is discussed here.

In February 2006, the Pressurized Water Reactor Owners Group (PWROG) issuedWCAP-16530-NP, Revision 0, "Evaluation of Post-Accident Chemical Effects inContainment Sump Fluids in Support of GSI-191." This WCAP provided a chemicalmodel that estimated the type and amounts of chemical precipitates that may beformed in a post-LOCA environment using plant specific containment materialsinventories and environmental conditions (sump and atmosphere temperatures).Throughout 2006, Duke used this chemical model to ascertain the boundingamounts of any precipitates that may form in a thirty day period (the designed ECCSmission time).

Page 15 of 38



Laboratory Aluminum Release Rate TestingDuke performed aluminum corrosion and aluminum release rate testing internally toexpand the PWROG test database. This testing has resulted in a betterunderstanding of the materials' performance in the expected post-LOCAcontainment sump pool environments at the Duke plants (McGuire, Catawba, andOconee Nuclear Stations).

WCAP-16530-NP, Revision 0 provided aluminum corrosion and release rate dataover short time spans (1.5 hours) and a wide pH range. The additional corrosionand release rate tests performed by Duke Energy provided data over longer timespans and in environments more consistent with the post-LOCA sump poolchemistry at McGuire and Catawba.

For McGuire and Catawba, the aluminum release rate algorithm located in theWCAP-1 6530-NP spreadsheet was modified based on this testing. Compared to theWCAP algorithm, the Duke algorithm estimated lower aluminum releases at low pHand slightly higher aluminum releases at high pH. For the predicted post-LOCAenvironment for McGuire and Catawba, the Duke algorithm results in aconservatively increased mass of aluminum released when compared to the WCAPresults.

Vertical Loop TestingTo better understand the sensitivity of the various input parameters used in thePWROG chemical model on the amount of possible chemical precipitates predictedand to gain insight into the behavior of various aluminum chemical species and theireffect on pressure drop across a fiberglass insulation bed deposited on arepresentative strainer, Duke Energy constructed a Vertical Test Loop assembly andconducted a series of tests. The loop was completed in late 2006 and a total offourteen tests with chemical additions were performed.

The Vertical Test Loop assembly which addressed both Catawba and McGuire wasconstructed using a flat plate strainer, and representative amounts of pre-treatedfiber insulation were used to form a bed on the strainer. Tests were performed usingpredicted site specific chemistry and with additions of either sodium aluminumsilicate particulate or soluble aluminum. Chemical loadings were based on modelpredictions using the PWROG chemical model provided by WCAP-1 6530-NP,modified with the Duke aluminum release rate algorithm, to estimate the amount ofchemicals released and possible precipitates that might form subsequent to aLBLOCA. Use of this aluminum release rate algorithm was conservative, because itresults in higher releases at the McGuire/Catawba estimated post-LOCA ECCScontainment sump pool pH than those resulting from the WCAP algorithm with thesame inputs. Flow velocities were conservatively based on Maximum Safeguardsconditions. The Catawba maximum volume, Minimum Safeguards temperatureprofile scenario resulted in the highest aluminum release and was used as the

Page 16 of 38



bounding case for both McGuire and Catawba. As a result of the test loop volume tostrainer area ratio being less than the plant ratio, the test loop concentrations of bothsilica and aluminum were considerably higher than actual concentrations predictedin the plant. This provided additional conservatism to the Vertical Test Loop results.

Request for Additional Information 11Provide a detailed description of any testing that has been or will beperformed as part of a plant-specific chemical effects assessment. Identify thevendor, if applicable, that will be performing the testing. Identify theenvironment (e.g., borated water at pH 9, deionized water, tap water) and testtemperature for any plant-specific head loss or transport tests, Discuss howany differences between these test environments and your plant containmentpool conditions could affect the behavior of chemical surrogates. Discuss thecriteria that will be used to demonstrate that chemical surrogates produced fortesting (e.g., head loss, flume) behave in a similar manner physically andchemically as in the ICET environment and plant containment poolenvironment.

McGuire Response:General Note:In order to address specific ECCS head loss/NPSH issues, Duke designed acomprehensive Integrated Prototype Test (IPT) to assess chemical effects on thepostulated post-LOCA debris bed at McGuire and Catawba. Bench-top tests(including laboratory tests and a 30-day Vertical Loop Test) were also performed aspart of an aluminum release rate testing program, in order to provide insights for theIntegrated Prototype Test parameters. The bench-top testing program that led to thedevelopment of the IPT and to the refinement of its input parameters is discussed inthe response to RAI #10 of Enclosure 1. The design and set-up of the IPT arediscussed in detail here. IPT results will be address as stated in the response to RAI12 of Enclosure 1.

Duke's strategy to evaluate chemical effects on the modified ECCS sump strainerhead loss employs an Integrated Prototype Test (IPT) designed to simulate thepredicted comprehensive challenge to the McGuire and Catawba ECCS Strainermodules (top-hats) in the post-LOCA containment pool. The IPT, performed by WyleLaboratories at their Huntsville, Alabama facility in fall 2007, combined the physicaland chemical characteristics expected in the post-accident environment just prior toECCS sump pool recirculation, and then challenged a prototype strainer top-hat inthe recirculating pool for the-full 30-day ECCS mission time while representativechemical effects were introduced.

Page 17 of 38

/



Integrated Prototype Test Setup

The test was conducted in a tank with a prototypical top-hat module mountedhorizontally. To closely match the interaction of the top-hat with the surroundingstrainer assemblies, vertical walls were placed in close proximity to the top-hatperforated plate on both sides. The module was also positioned so that the bottomedge of the lower base plate is in close proximity to the floor. The testing mimickedpost-LOCA containment pool conditions, including a representative flow rate ofborated water buffered with sodium tetraborate. A decreasing temperature profilewas also followed. Debris sources predicted to arrive at the strainer, includingNUKON®fiberglass, a failed coatings surrogate, and a dirt/dust surrogate, wereadded to the tank and allowed to accumulate on the top-hat module. In addition, anamount of NUKON® fiberglass insulation predicted to not physically transport to thetop-hat was submerged in the system fluid so it is available to react chemically withinthe pool. To represent the sprayed and submerged condition of containmentmaterials, a solution of aluminum nitrate is metered into the system over timeaccording to a predicted concentration profile. The test continued for 30 days, whiledebris bed head loss, flow rate, temperature, and pool pH were monitored andrecorded.

A 36-inch long prototype top-hat module was utilized in the IPT test loop. Thislength is representative of the population of top-hat modules available on theMcGuire Unit 1 and 2 modified strainers (24 inch, 30 inch, 36 inch and 45 inch) anddimensionally similar to the population of top-hat modules available on the CatawbaUnit 1 and 2 modified strainers. All other top-hat parameters (e.g., base plate, meshsize, diameter) on the test prototype module are identical to those installed in theMcGuire and Catawba ECCS sumps, including the bypass eliminator feature.

Page 18 of 38



Figure 11-1 below shows the physical IPT rig setup.

Figure 11-1

Integrated Prototype Test Setup

Test Environment

Representative Sump Pool Properties (boron, pH, buffering agent)

Boron and pHThe system boron concentration was initially 1730 ppm (+/- 300 ppm), addedas boric acid to demineralized water. The boron concentration wasapproximately 2400 ppm (+/- 500 ppm) after pH adjustment with sodiumtetraborate.

Page 19 of 38

Enclosure IResponses to Staff Request for Additional Information Identified on February 9, 2006


Buffering Agent

The pH was adjusted with sodium tetraborate to 7.9 (as measured at 250C).The pH remained between 7.8 and 8.0 during the test period. Sodiumhydroxide and nitric acid were used to maintain pH within the specification.

Representative sump pool temperature profile

In general, the Catawba post-LOCA minimum safeguards scenario has the highesttemperatures, and the McGuire post-LOCA maximum safeguards scenario has thelowest temperatures. The highest temperature profile was simulated initially, and thelowest temperature profile simulated during the latter part of the test. Thisconservative approach ensures a bounding sump pool condition for potentialchemical precipitates.

Figure 11-2 shows the expected temperature profiles for various post-LOCAscenarios at McGuire and Catawba during the ECCS mission time, and therepresentative IPT profile.

1*45 and CNS nost-LOCA Temoserahure260

230

220

210

200

190

100

ISO130

130

120

100

9010 100 1000 10000 100000 1000000 1000000C

10 ~ ~ ~ TD 10000 10)0

Fig-ure 11-2

Page 20 of 38



McGuire/Catawba Sump Pool Temperature Profiles

For the test, targeted pool temperature conditions (within +/- 5'F, with lineartransitions),were as follows:

* The test was to start at 190°F and remain constant for 30 minutes." The system was to decline to 1850F at 2 hours and remain constant until 10

hours.* The system was to decline to 1550F at 48 hours.* The system was to decline to 130°F at 192 hours.* The system was to decline to 11 0°F at 576 hours.* The system was to decline gradually to 90°F at 648 hours and remain constant

for the remainder of the test.

Representative approach velocity

An approach velocity of 0.0275 ft/sec was used, equating to a test flow rate throughthe top-hat module of approximately 114 gpm (+/- 5 gpm); this was above theexpected maximum safeguards average approach velocity to the McGuire andCatawba ECCS sump strainers.

Conventional Debris

General Note:The conventional debris quantities (fiber and coatings) used in the design of the IPTare refined quantities. Details regarding the refined quantities can be found inSections 3(b) and 3(h) of Enclosure 2.

Representative debris loading (fiber)

The fiber debris load designed to challenge the top-hat in the IPT test loop reflectedthe postulated loading generated from a large break LOCA located on the ReactorCoolant System B Loop Hot Leg at McGuire. This bounds the postulated loadinggenerated from the limiting large break LOCA located on the Reactor CoolantSystem B Loop Crossover Leg at Catawba. Nukon® fiber insulation was used torepresent fiber insulation systems for the IPT. The fiber debris load thickness for theIPT was conservatively approximated at 1.75 inches. This value is representative ofthe expected post-LOCA conditions are McGuire and bounding for Catawba. Also,to be representative of the expected post-LOCA condition at each plant, theinsulation was in a shredded form and baked to remove any organic binders.

In addition to the fiber debris load expected at the strainer top-hat, there was anadditional amount of fibrous debris expected to transport to the containment sumppool and submerge, but not make it to the strainer modules. This additional fibrous

Page 21 of 38



debris was therefore available to react with the containment sump pool fluid but notcollect on the strainer. The potential chemical interaction with this material wasaccommodated in the IPT via the immersion of more Nukon® fiber to the test pool ina proportioned amount appropriate to the refined debris analysis.

Representative particulate loading (dust, dirt, failed coatings)

The surrogate material for latent dirt and dust is a material blend of silica sandrepresentative of PWR latent dirt/dust particles. The size distribution of the sand wasprepared consistent with the latent dirt/dust distribution provided in the NRC SER ofNEI 04-07.

The failed coatings debris surrogate material was selected based on chemicalreactivity and a comparison of microscopic densities. Epoxy and alkyd coatingsdensities at plants range from 94 lb/ft3 to 98 lb/ft3 per NEI 04-07 guidance. Thesurrogate used for epoxy and alkyd coatings was silica flour, which has a materialspecific gravity of 2.65 (microscopic density of 165.4 lb/ft3).

The critical parameter for selecting the surrogate material is the volume of thematerial in the debris mix. The particulate material occupies a certain volume in thefibrous debris space that results in increasing resistance to flow, and thereforehigher head loss. The surrogate material volume was .adjusted to match the volumeof the failed coatings particulate when it is less dense than the surrogate. Theparticle size for all failed coatings (epoxy, alkyd, and inorganic zinc) is assumed tobe 10 microns per NEI 04-07 guidance. The surrogate materials were a sphericalparticulate, where 99% is less than 45 microns in diameter and 69% is less than 10microns in diameter.

Chemical debris addition (calcium, silica)Ilnjection of dissolved aluminum

Chemical debris in the post-LOCA environment for McGuire and Catawba will largelyinclude dissolved aluminum and silica. While no calcium precipitates are predicted toform in the containment sump pool, calcium chloride is added to the IPT pool toachieve a representative I conditions. calcium concentration that mimics expectedcontainment poo

CalciumThe IPT calcium concentrations simulate the highest estimated plantreleases. Other particulate additions (fiberglass, surrogate debris materials,etc.) were taken into account to obtain a final solution calcium concentrationof approximately 7-10 ppm.

SiliconThe scaled volume of non-transported NUKON® was submerged in the testpool fluid while preventing it from reaching the top-hat module. Dissolution ofthe submerged NUKONO by the pool chemistry provides the majority of

Page 22 of 38



dissolved silicon as predicted by the WCAP methodology. Due to temperaturelimitations of the tank, the test was conducted at a lower temperature for thefirst hour. The silicon predicted by the WCAP methodology to be released inthis period was added as sodium silicate. The dissolved aluminum and siliconmay react with sodium from the pool buffering agent to form sodiumaluminum silicate, a potential precipitate. The final total silicon concentrationfrom all sources in the IPT pool was approximately 30 ppm to bound thehighest predicted concentration in the post-LOCA containment sump pool.

Dissolved AluminumAn aqueous solution of aluminum nitrate is metered into the system based onthe aluminum release profile predicted by the WCAP-16530-NP model,assuming minimum safeguards and the McGuire (MNS) and Catawba (CNS)aluminum release rate algorithm determined from the Duke bench-top testingprogram. To the extent reasonable, a scaled Catawba minimum safeguardsrelease rate for the IPT was simulated, since it is bounding for both plants. Inaddition, to demonstrate stability, the IPT includes a run period with noaluminum injection at the end of the test.

Figure 11-3 shows the expected aluminum release rates at McGuire andCatawba for various post-accident scenarios during the ECCS mission time.

12

10

(P

0:

2

0 100 200 300 400 500 600 700

Time (hrs)

Figure 11-3McGuirelCatawba Post-LOCA Aluminum Release Rates

Page 23 of 38


McGuire Nuclear Station Units, 1 and 2Generic Letter 2004-02

For the IPT, the target aluminum injection profile was to follow the schedule shownin Table 11-1 below. The target quantities were to be achieved by slowly injectingthe solution to minimize the local concentration at the injection location.

Table 11-iAluminum Injection Schedule

Cumulative Volume ofAluminum Nitrate as

Start of End of 0.167% Rate ofInterval Interval AI(NO3)3.9H20 Addition

(hrs) (hrs) Solution (gal) (gph)0 8 10.1 1.278 18 18.4 0.8318 24 21.7 0.5524 48 31.7 0.4248 96 43.8 0.26

96 168 56.1 0.18168 288 71.2 0.13

288 504 92.8 0.10504 672 98.1 0.04

For reference, Table 11-2 shows the final estimated releases and concentrations forthe McGuire and Catawba post-LOCA ECCS sump pool assuming minimumsafeguards temperatures, the Duke aluminum release rate algorithm, estimatedaluminum surface areas and wetted fiberglass volume estimates.

Page'24 of 38



Table 11-2McGuire and Catawba Final Estimated Releases and Concentrations

CNS max. vol. CNS min. vol. MNS max. vol. MNS min. vol.Ca Release (kg) 14.72 14.72 14.72 14.72

Si Release (kg) 80.51 52.44 52.21 34.25

Al Release (kg) 11.38 10.99 7.89 7.68Ca Concentration (ppm) 4.5 8.3 4.5 8.2

Si Concentration (ppm) 24.4 29.4 15.8 19.2Al Concentration (ppm) 3.4 6.2 2.4 4.3Al/Gross Screen area(g Alift2 4.74 1_4.58 1 4.64 1 4.52

Assumptions for this table: Minimum volume = 1784857.2 kg waterMaximum volume =3300017.2 kg waterCNS Final Gross Screen Area = 2400 ft2

CNS Final Net Screen Area = 1772 ft2

MNS Final Gross Screen Area = 1700 ft2

MNS Final Net Screen Area = 1317 ft2

CNS reduced wetted fiberglass estimate = 625 ft3

MNS reduced wetted fiberglass estimate = 625 ft3

CNS final submerged aluminum = 563.2 ft2

CNS final sprayed aluminum = 140.1 ft2

MNS final submerged aluminum = 530 ft2

MNS final sprayed aluminum = 736 ft2

Page 25 of 38



Request for Additional Information 12For your plant-specific environment, provide the maximum projected headloss resulting from chemical effects (a) within the first day following a LOCA,and (b) during the entire ECCS recirculation mission time. If the response tothis question will be based on testing that is either planned or in progress,provide an estimated date for providing this information to the NRC.

McGuire Response:The response to RAI #12 will be based on the results of the Integrated PrototypeTest (IPT). Maximum projected head loss, limiting NPSH margins and all supportinginformation regarding refinements to the initial evaluations will be provided by April30, 2008 as described in the December 28, 2007 extension request approval.

Request for Additional Information 13Results from the ICET #1 environment and the ICET #5 environment showedchemical products appeared to form as the test solution cooled from theconstant 140°F test temperature. Discuss how these results are beingconsidered in your evaluation of chemical effects and downstream effects.

McGuire Response:As described in the response to RAI #10 of Enclosure 1, aluminum solubilityconditions were evaluated by Duke Energy via bench-top testing. As part of thatevaluation, the final temperature during the 30-day Vertical Loop Test was less thanthe lowest predicted McGuire ECCS post-LOCA containment pool temperature at 30days.

As described in the response to RAI #11 of Enclosure 1, these Vertical Loop Testevaluations were incorporated into the test plan for the Integrated Prototype Test(IPT) for chemical effects. The test plan simulated the full range of expected post-accident ECCS sump temperatures at McGuire. As in the Vertical Loop Test, theIPT final temperature was less than the lowest predicted ECCS sump temperaturefor McGuire at 30 days. To achieve this, the temperature of the IPT was reduced to90°F during the latter part of the test. This test evaluated aluminum solubility, usingspecific chemistry and environmental parameters for McGuire Nuclear Station. TheIPT results will be used to assess the total predicted post-LOCA head loss throughthe modified McGuire ECCS sump strainers, including chemical effects.

Downstream chemical effects are addressed in Section 3(o) of Enclosure 2.

Page 26 of 38



Request for Additional Information 25Describe how your coatings assessment was used to identify degradedqualified/acceptable coatings and determine the amount of debris that willresult from these coatings. This should include how the assessmenttechnique(s) demonstrates that qualified/acceptable coatings remain incompliance with plant licensing requirements for design-basis accident (DBA)performance. If current examination techniques cannot demonstrate thecoatings' ability to meet plant licensing requirements for DBA performance,licensees should describe an augmented testing and inspection program thatprovides assurance that the qualified/acceptable coatings continue to meetDBA performance requirements. Alternately, assume all containment coatingsfail and describe the potential for this debris to transport to the sump.

McGuire Response:The comprehensive Duke Energy Corporation Containment Coatings AssessmentProgram in effect at McGuire Nuclear Station is used to identify degradedqualified/acceptable coatings and determine the amount of debris that will resultfrom these coatings. This program also ensures that qualified/acceptable coatingsremain in compliance with plant licensing requirements for design-basis accident(DBA) performance. If, after identification, degraded qualified/acceptable coatingswill be left in place during plant operation, the degraded qualified/acceptablecoatings are assumed to fail and to be available for transport to the ECCS sump.

Insights on the Containment Coatings Assessment ProgramAs originally discussed in Duke Energy Corporation's McGuire Nuclear Stationresponse dated November 11, 1998 to Nuclear regulatory Commission (NRC)Generic Letter 98-04, "Potential for Degradation of the Emergency Core CoolingSystem and the Containment Spray System After a Loss-of-Coolant AccidentBecause of Construction and Protective Coating Deficiencies and Foreign Material inContainment", a comprehensive program is in place at McGuire Nuclear Station forassessing and documenting the condition of qualified/acceptable coatings in primarycontainment. This program generates data which is used to schedulequalified/acceptable coating maintenance to ensure that qualified/acceptable primarycontainment coatings will not fail (detach) during normal and accident conditions andthus not contribute to the Emergency Core Cooling System (ECCS) debris sourceterm.

The Containment Coating Assessment Program is controlled through a NuclearGeneration Department level document. This guidance document specifies detailsfor assessing and developing the condition of all coatings, includingqualified/acceptable coatings, located in the McGuire Nuclear Station primarycontainments. The requirements of the Containment Coating Assessment programare procedurally implemented at McGuire.

Page 27 of 38



A primary containment coatings condition assessment is conducted during eachrefueling outage. Visual inspections are conducted and documented by ANSIN45.2.6 Level II personnel and/or personnel who have demonstrated overalltechnical knowledge of coatings. The resultant data is reviewed by the site CoatingSpecialist and is used to facilitate proper planning and prioritization of coatingsmaintenance as needed to maintain the integrity of qualified/acceptable primarycontainment coating systems.

The guidance provided in ASTM D5163, "Standard Guide for EstablishingProcedures to Monitor the Performance of. Coating Service Level I Coating Systemsin an Operating Nuclear Power Plant," and EPRI Report 109937 "Guideline onNuclear Safety-Related Coatings: Revision 1" (November 2001) is incorporated inthe McGuire Nuclear Station primary containment coatings condition assessmentprogram. The primary containment coating condition assessment protocol consistsof a visual inspection of all readily accessible coated areas by qualified personnel.The use of visual inspection by qualified personnel for containment coatingassessment has been validated by. the recently-issued EPRI Report 1014883 "PlantSupport Engineering: Adhesion Testing of Nuclear Coating Service Level 1Coatings" (August 2007).

When degraded coatings are visually identified, the affected areas are documentedin accordance with plant procedures. Additional nondestructive and/or destructiveexaminations are conducted as appropriate to define the extent of the degradedcoatings and to enable disposition of the coating deficiency. The guidance containedin EPRI Report 109937 is used as appropriate to disposition areas of degradedcoatings when discovered, including:

1. performance of additional in situ and/or laboratory testing of degraded coatings,2. removal and replacement of degraded coatings,3. repairing degraded coatings,4. mitigation of accident consequences related to failure of degraded coatings,5. leaving in place based on evaluation of effects of failure (detachment) of the

degraded coating on ECCS system performance, and/or,6. upgrading of indeterminate coatings.

The following industry technical documents are used as appropriate in determiningthe physical characteristics of debris resulting from any degraded coatings identifiedduring primary containment coatings condition assessments:

1. "Analysis of Pressurized Water Reactor Unqualified Original EquipmentManufacturer Coatings", EPRI Report 1009750, March 2005.

Page 28 of 38



2. "Design Basis Accident Testing of Pressurized Water Reactor UnqualifiedOriginal Equipment Manufacturer Coatings", EPRI Report 1011753, September2005.

3. Keeler & Long Report No. 06-0413, "Design Basis Accident Testing of CoatingSamples from Unit 1 Containment, TXU Comanche Peak SES."

Request for Additional Information 30The NRC staffs safety evaluation (SE) addresses two distinct scenarios forformation of a fiber bed on the sump screen surface. For a thin bed case, theSE states that all coatings debris should be treated as particulate andassumes 100% transport to the sump screen. For the case in which no thinbed is formed, the staffs SE states that the coatings debris should be sizedbased on plant-specific analyses for debris generated from within the ZOI andfrom outside the ZOI, or that a default chip size equivalent to the area of thesump screen openings should be used (Section 3.4.3.6). Describe how yourcoatings debris characteristics are modeled to account for your plant-specificfiber bed (i.e. thin bed or no thin bed). If your analysis considers both a thinbed and a non-thin bed case, discuss the coatings debris characteristicsassumed for each case. If your analysis deviates from the coatings debrischaracteristics described in the staff-approved methodology, providejustification to support your assumptions.

McGuire Response:The modified McGuire ECCS sump strainer utilizes an array of strainer modules(top-hats) that do not exhibit thin-bed formation.

The coatings debris analysis for McGuire followed the staff-approved methodologyfor the non-thin bed case. The exception is that a ZOI of 5D is assumed in lieu of the10D ZOI prescribed in the SER, based on the results of specific testing performedunder WCAP-16568-P.

For post-LOCA debris generation analyses, qualified coatings within the 5D ZOI atthe limiting High Energy Line Break location were postulated to fail, as well as allunqualified coatings within the containment building. Qualified coatings within the 5DZOI and all unqualified coatings are assumed to fail as 10 micron spheres, 100% ofwhich transport to the ECCS sump.

A detailed discussion of coatings debris characteristics and analytical assumptionscan be found in Section 3(h) of Enclosure 2.

Page 29 of 38



Request for Additional Information 31Was/will "leak before break" be used to analyze the potential jet impingementloads on the new ECCS sump screen?

McGuire Response:The application of "leak before break" (LBB) technology was approved for McGuireUnits 1 and 2 primary coolant loops by NRC letter dated May 8, 1986. As describedin Section 3.6.2.1.1 of the McGuire UFSAR, LBB was used to eliminate breakspostulated in the original design of the Reactor Coolant System primary loop piping.Postulated high energy pipe breaks which could potentially interact with the modifiedECCS containment sump strainer were evaluated in accordance with McGuire'scurrent licensing basis as follows:

a. High Energy Pipe Rupture composite drawings were reviewed to identify anypostulated breaks in close proximity to the modified ECCS sump strainerassembly and associated enclosure.

b. Postulated breaks were evaluated to determine if interaction existed with themodified ECCS containment sump strainer and associated enclosure by being,within the target zone of pipe whip or jet impingement.

Per the above methodology, one interaction per Unit at McGuire was identifiedbased upon the new locations of the modified strainer assemblies.

Regulatory commitments were made to install pipe rupture restraints/jet barriers onthe Residual Heat Removal System of each Unit to address this interaction. Therupture restraint/jet barrier has been installed on Unit 1 (1EOC18 outage, spring2007), and the rupture restraint/jet barrier will be installed on Unit 2 prior tobeginning cycle 19 (2EOC18 outage, spring 2008).

LBB methodology was only used for evaluation of pipe whip, jet impingement anddynamic loads on the modified McGuire ECCS strainer. LBB methodology was notused in the GSI-191 determination of debris generated as a result of a LOCA. A fullyoffset, double-ended guillotine break of the primary coolant loop was used for debrissource term. The break location chosen was bounding for debris source term.Further details regarding the debris generation evaluation are located in Section 3(b)of Enclosure 2.,

Page 30 of 38


McGuire Nuclear Station Units 1 and 2Generic Letter'2004-02

Request for Additional Information 32You indicated that you would be evaluating downstream effects in accordancewith WCAP 16406-P. The NRC is currently involved in discussions with theWestinghouse Owner's Group (WOG) to address questions/concernsregarding this WCAP on a generic basis, and some of these discussions mayresolve issues related to your particular station. The following issues have thepotential for generic resolution; however, if a generic resolution cannot beobtained, plant-specific resolution will be required. As such, formal RAIs willnot be issued on these topics at this time, but may be needed in the future. Itis expected that your final evaluation response will specifically address thoseportions of the WCAP used, their applicability, and exceptions taken to theWCAP. For your information, topics under ongoing discussion include:

a. Wear rates of pump-wetted materials and the effect of wear on componentoperation

b. Settling of debris in low flow areas downstream of the strainer or creditfor filtering leading to a change in fluid composition

c. Volume of debris injected into the reactor vessel and core regiond. Debris types and propertiese. Contribution of in-vessel velocity profile to the formation of a debris bed

or clogf. Fluid and metal component temperature impactg. Gravitational and temperature gradientsh. Debris and boron precipitation effectsi. ECCS injection pathsj. Core bypass design featuresk. Radiation and chemical considerationsI. Debris adhesion to solid surfacesm. Thermodynamic properties of coolant

McGuire Response:The downstream effects issues identified in this 2006 RAI are addressed via theNRC's "Revised Content Guide for Generic Letter 2004-02 SupplementalResponses", dated November 2007. McGuire responses to the related ContentGuide issues are located in Section 3(m), Section 3(n), and Section 3(o) ofEnclosure 2.

Page 31 of 38



Request for Additional Information 33Your response to GL 2004-02 question ( d )(viii) indicated that an activestrainer design will not be used, but does not mention any consideration ofany other active approaches (i.e., backflushing). Was an active approachconsidered as a potential strategy or backup for addressing any issues?

McGuire Response:The use of backflushing (or other active mitigative strategies) was not consideredfeasible for the McGuire modified ECCS strainer design.

Request for Additional Information 34

McGuire stated that the containment walkdown for Unit 1 will be completed inaccordance with Nuclear Energy Institute (NEI) 02-01 during the fall 2005outage. Please discuss the plans to incorporate the results of this futurecontainment walkdown into the sump design analyses.

McGuire Response:The McGuire ECCS sump strainer GSI-191 Baseline Analysis used the Unit 2containment walkdown results, since at the time the Unit 1 containment walkdownshad not been completed. It was assumed that the Unit 2 debris source terms wouldbound the Unit 1 debris source terms for the purposes of the initial strainer sizingcalculations.

The McGuire Unit 1 containment walkdown was completed as scheduled during thefall 2005 outage in accordance with NEI 02-01 requirements. The walkdownfindings confirmed that the debris sources identified by the McGuire Unit 2containment walkdown bound the Unit 1 debris sources with two exceptions:

a. The insulation on the Unit 1 reactor coolant pumps is Nukon® (a fibrousinsulation), while the Unit 2 reactor coolant pumps have the original reflectivemetal insulation (RMI) installed.

The debris source term introduced by the additional Nukon® insulation on the Unit 1reactor coolant pumps was incorporated into the McGuire Unit 1 ECCS sumpstrainer design analysis.

b. The latent debris load (dust, dirt, and lint) in Unit 1 was estimated to be higherthan that estimated in Unit 2, based on specific sampling.

The increased amount of latent debris (dust, dirt, and lint) estimated for McGuireUnit 1 is accommodated by the conservative initial source term assumption of 200Ibm in the ECCS sump strainer design analysis. Both the Unit 1 and the Unit 2

Page 32 of 38



latent debris containment walkdown estimates for latent debris are well below thisassumed value.

The modified ECCS sump strainer design for McGuire Units 1 and 2 incorporatedthese values, as applicable, which were refined following the Unit 1 walkdown.

Request for Additional Information 35You stated that Microtherm® insulation (currently installed on portions of thereactor vessel heads) will be replaced, and that this replacement will reducethe postulated post-accident debris loading on the sump strainer. Pleasediscuss the insulation material that will replace the Microtherm® insulation,including debris generation and characteristics parameters. Has the newinsulation been evaluated in the debris generation, transport, head lossanalyses and other sump design analyses?

McGuire Response:Microtherm® insulation previously installed on portions of the McGuire Unit 1 andUnit 2 reactor vessel heads was removed and replaced with Reflective MetalInsulation (RMI). The reactor vessel head is not in the area of the limiting break. Inaddition, debris transport calculations that have been completed for McGuire Units 1and 2 show that RMI (for the limiting break location) will not transport to thecontainment sump. Debris transport for the limiting break is described in detail in3(e), Table 3E6-1 of Enclosure 2.

Request for Additional Information 36You did not provide information on the details of the break selection, ZOI anddebris characteristics evaluations other than to state that the Nuclear EnergyInstitute (NEI) and SE methodology were applied. Please provide a descriptionof the methodologies applied in these evaluations and include a discussion ofthe technical justification for deviations from the SE-approved methodology.

McGuire Response:A detailed discussion of the methodologies for Break Selection, Zone of Influence(ZOI), and Debris Characteristics evaluations, as they apply to the modified McGuireECCS strainer design, are located in Enclosure 2 of this submittal. The specificsections are identified below.

Break Selection Evaluation Methodology

0 Enclosure 2, Section 3(a)

Page 33 of 38



Zone of Influence (ZOI) Evaluation Methodologies

" Enclosure.2, Section 3(b) for insulation" Enclosure 2, Section 3(h) for coatings

Debris Characteristics Evaluation Methodology

* Enclosure 2, Section 3(C) for fiber and particulate debris* Enclosure 2, Section 3(d) for latent debris

Request for Additional Information 37Has debris settling upstream of the sump strainer (i.e., the near-field effect)been credited or will it be credited in testing used to support the sizing oranalytical design basis of the proposed replacement strainers? In the casethat settling was credited for either of these purposes, estimate the fraction ofdebris that settled and describe the analyses that were performed to correlatethe scaled flow conditions and any surrogate debris in the test flume with theactual flow conditions and debris types in the plant's containment pool.

McGuire Response:Upstream debris settling due to the "near-field effect" is not credited in the head losstesting or in the analytical design basis for the sizing of the McGuire modified ECCSsump strainers. The debris transport calculation (and so the strainer design basis)assumes that 100% of the particulate debris (failed coatings) and latent debris (dust,dirt, and lint) will challenge the strainer after the limiting break. The debris transportfractions for destroyed insulation were determined based on a computational fluiddynamics model.

Page 34 of 38

U



Request for Additional Information 38Are there any vents or other penetrations through the strainer control surfaceswhich connect the volume internal to the strainer to the containmentatmosphere above the containment minimum water level? In this case,dependent upon the containment pool height and strainer and sumpgeometries, the presence of the vent line or penetration could prevent a waterseal over the entire strainer surface from ever forming; or else this seal couldbe lost once the head loss across the debris bed exceeds a certain criterion,such as the submergence depth of the vent line or penetration. According toAppendix A to Regulatory Guide 1.82, Revision 3, without a water seal acrossthe entire strainer surface, the strainer should not be considered to be "fullysubmerged," Therefore, if applicable, explain what sump strainer failurecriteria are being applied for the "vented sump" scenario described above.

McGuire Response:There are no vents or other penetrations through the modified McGuire ECCSstrainer connecting the interior of the strainer to the containment atmosphere abovethe containment minimum water level. The McGuire strainer is designed to be fullysubmerged, and as discussed in the response to RAI #41 of Enclosure 1, is fullysubmerged even in the bounding SBLOCA scenario.

Request for Additional Information 39What is the minimum strainer submergence during the postulated LOCA? Atthe time that the re-circulation starts, most of the strainer surface is expectedto be clean, and the strainer surface close to the pump suction line mayexperience higher fluid flow than the rest of the strainer. Has any analysisbeen done to evaluate the possibility of vortex formation close to the pumpsuction line and possible air ingestion into the ECCS pumps? In addition, hasany analysis or test been performed to evaluate the possible accumulation ofbuoyant debris on top of the strainer, which may cause the formation of an airflow path directly through the strainer surface and reduce the effectiveness ofthe strainer?

McGuire Response:Minimum SubmergenceAs discussed in the response to RAI #41 of Enclosure 1, the minimum McGuireECCS sump strainer submergence during the limiting SBLOCA containment sumppool inventory scenario is at least 2 inches.

Vortex Formation EvaluationWith an initially clean ECCS sump strainer surface, approach velocities for the top-hat modules closest to the pump suction line are expected to be higher than the

Page 35 of 38



McGuire predicted nominal approach velocity by approximately a factor of two. Afull-scale 36 inch long top-hat module was used to test for vortex formation atvarious approach velocities, while conservatively maintaining the water level only 3inches above the top surface of the top-hat perforated plate (the expected minimumwater level above this surface is about 4 inches). The approach velocity through theclean module was increased until an air-entraining vortex was formed,. and then thevortex suppressor grating was placed into a position above the top-hat module. Theair-entraining vortices that formed at higher approach velocities were eliminated bythe vortex suppressor grating in each case, and only minor surface dimplingremained. No vortices were observed at lower approach velocities. Vortexsuppression testing is discussed in detail in Section 3(f) of Enclosure 2, anddemonstrates that the McGuire ECCS sump strainers are not susceptible to air-entraining vortex formation.

Accumulated Buoyant DebrisThe McGuire ECCS strainer is designed to be fully submerged and un-vented, asidentified in the response to RAI #38 of Enclosure 1. The only portion of thepredicted post-LOCA debris load that could potentially remain buoyant is the lowdensity fiberglass insulation (LDFG), and industry testing has shown that LDFGinsulation debris becomes saturated and sinks very quickly in hot water, while intactjacket-covered "pillows" can remain afloat. In the unlikely event that some largeintact pieces survive the blowdown and transport to the pipechase through the cranewall penetrations, the fully submerged Pipechase Vortex Suppression Rack will besufficient to preclude any opportunity for an artificial vent to form between the top-hats and the surface of the water.

Request for Additional Information 40Please provide a detailed description of the analyses/testing performed toevaluate the new strainer head loss.

McGuire Response:Detailed discussion regarding the analyses/testing performed to evaluate themodified McGuire ECCS strainer head loss is located in Section 3(f) and Section3(o) of Enclosure 2.

Page 36 of 38



Request for Additional Information 41Describe, in detail, the analysis performed to determine the minimum waterlevel, and explain how the water source from the ice condenser is determined.

McGuire Response:The limiting analytical case for minimum ECCS sump level at McGuire can becharacterized as a small break LOCA (SBLOCA) during which Containment Spraydoes not actuate and there is no water source contribution from ice melt. In addition,this containment analysis conservatively accounts for potentially diverted ECCSinjection inventory.

The analysis assumes a small break of indeterminate size which fills up the incoreinstrumentation room (located below the reactor vessel), but has insufficient energyto cause the Containment Spray system to actuate. No ice melt is credited in thisanalysis. Credited water for this specific accident includes the TechnicalSpecification minimum inventory from the Refueling Water Storage Tank (FWST),and the FWST low-low level setpoint is conservatively error-adjusted upward tominimize the usable FWST volume. The following ECCS sump inventory penalties(lost water sources) are applied in this analysis:

* Reactor Coolant System shrinkage

* Incore instrumentation room diversion

" Volume Control Tank diversion

" Pressurizer Relief Tank diversion

* Lower containment ventilation system condensation diversion (loss of lowercontainment condensate through drain pans and drain lines)

For this SBLOCA scenario, a sump pool level of 36 inches above the sump floor iscalculated. This corresponds to a water level which is at least 2.inches above thevortex suppressor of the modified McGuire sump strainers. The ECCS sump levelswitch set point is selected such that the level alarm will indicate sufficient level forECCS pump alignment to the sump under the limiting case. This is the minimumcalculated water level for the bounding SBLOCA scenario, and confirms that themodified ECCS sump structure at McGuire is submerged when the FWST reachesthe low-low level setpoint.

For SBLOCA scenarios which automatically actuate the Containment Spray system,sump strainer submergence is also confirmed. Additional inventory penalties relatedto the use of containment spray are applied to account for holdup of water withinupper containment (beneath the elevation of the refueling canal drains, within

Page 37 of 38



Containment Spray piping, film layers on walls, etc.). Since any break size whichwould cause the Containment Spray system to automatically actuate (at acontainment pressure of 3 psig) would also lead to the opening of the Ice CondenserInlet Doors, an ice melt contribution is calculated for those events. Additionally, theinteraction of the sump level switches and the variation of the incore instrumentationroom volume as a function of time are considered.

Request for Additional Information 42Duke's September 2005 GL response stated that the design of the modifiedcontainment sump would accommodate the effects of debris loading asdetermined by the baseline evaluation, which was under review by Duke, andthe ongoing refined evaluation for Catawba and that the evaluations use theguidance of NEI 04-07, "Pressurized Water Reactor Sump PerformanceEvaluation Methodology, Revision 0," dated December 2004. Pleasesupplement your GL response after completing the review.

McGuire Response:A supplement to McGuire's GL 2004-02 response based on the NRC's guidance ofNovember 21, 2007 is included as Enclosure 2 of this submittal. McGuire will furthersupplement its response by April 30, 2008 as committed in its November 6, 2007letter.

Page 38 of 38

Enclosure 2

Generic Letter 2004-02 Supplemental Responses

Enclosure 2Information Addressing Issues Identified in Staff Content Guide for Generic Letter

2004-02 Supplemental ResponsesMcGuire Nuclear Station

Specific Guidance for Review Areas

1. Overall Compliance:Provide information requested in GL 2004-02 Requested Information Item2(a) regarding compliance with regulations.

GL 2004-02 Requested Information Item 2(a)Confirmation that the ECCS and CSS recirculation functions under debrisloading conditions are or will be in compliance with the regulatoryrequirements listed in the Applicable Regulatory Requirements section ofthis GL. This submittal should address the configuration of the plant thatwill exist once all modifications required for regulatory compliance havebeen made and this licensing basis has been updated to reflect the resultsof the analysis described above.

McGuire ResponseGL 2004-02 Requested Information Item 2(a) requests confirmation of threerelated items. The first item is confirmation that the ECCS and CSS recirculationfunctions under debris loaded conditions are in compliance with the regulatoryrequirements listed in the Applicable Regulatory Requirements section ofGL2004-02:

The status of McGuire Nuclear Station's compliance with the regulatoryrequirements includes:

McGuire Nuclear Station requested in a letter dated February 9, 2007, anextension to complete the Unit 2 ECCS sump strainer. McGuirerequested an extension to the spring 2008 refueling outage to completeinstallation of the Unit 2 ECCS Sump strainer. This extension wasapproved by the US NRC in a letter dated December 19, 2007.

McGuire Nuclear Station, Units 1 and 2 requested in a letter datedNovember 6, 2007, an extension to complete the chemical effects testingand update the associated reports and design documents. Any additionalor revised information resulting from the Integrated Performance Test(chemical effects testing) will be provided as an amended response byApril 30, 2008. The extension was approved by the staff in a letter datedDecember 28, 2007.

McGuire Nuclear Station anticipates being in full compliance with theapplicable regulatory requirements for long term core cooling, containmentheat removal, and containment atmospheric cleanup by the requestedextension dates. The April 30, 2008 amended response will include thisconfirmation.

Page 1 of 86



The second confirmation item is to describe the configuration of the plant once allmodifications are made:

McGuire Nuclear Station will have installed modified ECCS strainers ineach Unit to address issues identified in GL 2004-02. The modified ECCSsump strainers will have increased the surface area from the original 135square feet, to approximately 1700 square feet. The strainer hole size isreduced from 0.206 inch (original) to less than 0.094 inch nominal (newstrainer). The McGuire ECCS sump design is described in detail in theMarch 8, 2007 License Amendment Request.

McGuire Nuclear Station has removed the Microtherm insulation,previously installed on the reactor vessel head on each Unit. Theinsulation was replaced with metal reflective insulation (RMI).

The third confirmation item is to describe the licensing basis of the plant once all

modifications are made:

The status of the licensing basis updates includes:

A license amendment request was submitted by Duke on March 8, 2007and supplemented March 27, April 13, and May 3 2007, to update thelicensing basis of McGuire Nuclear Station relative to the modified ECCSstrainer configuration. The purpose of this license amendment requestwas to revise the licensing commitments to Regulatory Guide 1.82, andrevise Technical Specification Surveillance Requirement 3.5.2.8. Thislicense amendment request was approved by amendments 240 (Unit 1)and 222 (Unit 2).

UFSAR changes will be made to update the licensing basis for otheraspects of GL 2004-02 to describe the revised debris loaded ECCS sumpstrainer license basis, including:

" Break Selection* Debris Generation* Latent Debris. Debris Transport* Head Loss* Additional Design Considerations

The UFSAR is submitted periodically to the USNRC. McGuire NuclearStation provides this required update 6 months after each Unit 2 refuelingoutage. Unit 2 has a Spring 2008 refueling outage where the sump

Page 2 of 86



strainer installation will be completed. Therefore, the update includingthese changes will be submitted by fall 2008.

2. General Description of and Schedule for Corrective Actions:Provide a general description of actions taken or planned, and dates foreach. For actions planned beyond December 31, 2007, reference approvedextension requests or explain how regulatory requirements will be met asper Requested Information Item 2(b). (Note: All requests for extensionshould be submitted to the NRC as soon as the need becomes clear,preferably not later than October 1, 2007.)

McGuire ResponseMcGuire Nuclear Station has used the guidance of NEI 04-07 to address ECCSsump performance. The analysis results required modifications to the ECCSsump strainers. The Unit 1 modification is complete, and the Unit 2 modificationwill be complete in the Spring 2008.

The following major activities have been completed in support of GL 2004-02:

" Baseline evaluation, performed by Enercon Services, Inc.* Refined evaluation using the guidance of NEI 04-07, completed by

Enercon Services, Inc.* Downstream effects evaluation using the WCAP-1 6406-P, Rev. 0

methodology.* Containment walkdowns using the guidance of NEI 02-01, "Condition

Assessment Guidelines: Debris Sources Inside PWR Containments"* The modification process and the plant labeling process have been

enhanced relative to GL 2004-02 controls.* Replacement of the Microtherm® insulation, previously installed on

portions of the reactor vessel heads, with RMI." Installation of a new ECCS sump strainer in Unit 1 (-1700 sq ft).* Installation of a new ECCS sump strainer in Unit 2 (phase I, -1000 sq ft)* Completion of the Integrated Prototype Test (Chemical Effects test)

The McGuire Unit 2 ECCS Sump strainer will be completed in the Spring 2008refueling outage. The only significant activity remaining is completion of theanalysis /report for the Integrated Prototype Test (chemical effects test) andincorporation of the results into the ECCS system NPSH calculations. Anyadditional or revised information resulting from the Integrated Performance Test(chemical effects testing) will be provided as an amended response by April 30,2008.

Page 3 of 86



As requested in the staff's extension approval letter dated December 28, 2007Duke will provide additional information related to the NRC staff-requestedevaluation of WCAP-1 6406, Revision 1 dated August 2007 in the April 30, 2008submittal.

Page 4 of 86



3(a) Break SelectionThe objective of the break selection process is to identify the break sizeand location that present the greatest challenge to post-accident sumpperformance.

3(a)(1) Describe and provide the basis for the break selection criteriaused in the evaluation.

McGuire Response:Break locations were selected for the breaks that produce the maximum amountof debris and also the worst combination of debris mixes with the possibility ofbeing transported to the ECCS sump strainer.

Additionally, breaks that might cause a "thin-bed" effect (i.e., a high volume ofparticulate debris on a thin fiber bed) were given consideration since these alsohave the potential to significantly impair sump strainer performance. Thefollowing break locations were analyzed for McGuire Units 1 and 2:

Break 1: Locations in the Reactor Coolant System (RCS) with the largestpotential for debris generation.

* Break 2: Locations with two or more different types of debris.* Break 3: Locations with the most direct path to the sump.* Break 4: Locations with the largest potential particulate to insulation ratio.* Break 5: Locations that would generate debris that could potentially form

a thin-bed.

Insights in the McGuire break selection process were gained from the NRC SERof NEI 04-07. The SER advocates break selection at 5 ft intervals along a pipe inquestion but clarifies that "the concept of equal increments is only a reminder tobe systematic and thorough." It further qualifies that recommendation by notingthat a more discrete approach driven by the comparison of debris source termand transport potential can be effective at placing postulated breaks.

The key difference between many breaks (especially large breaks) will not be theexact location along the pipe, but rather the envelope of containment materialtargets affected. A 17D ZOI for the Nukon® insulation (jacketed and unjacketed)used on RCS piping and components at McGuire is equivalent to a sphere withan approximate 40 ft radius, depending upon the size of the particular pipe break.A spherical ZOI of this size is bounded by structural barriers surrounding theRCS such as the reactor cavity, the crane wall, and the operating floor slabs.Also, due to the size of this ZOI, the specific location along a particular pipe haslittle if any impact on the amount of debris generated. Further, a reasonable

Page 5 of 86



determination of the most limiting location can be made by inspection of plantequipment drawings. Specific break locations can be selected by plotting theZOI along the RCS piping to maximize major targets that fall within the perimeterof the ZOI sphere.

Following initial break selection as described above which was used to designthe modified sump strainers, refinements in the break selection criteria were usedto develop sump strainer testing parameters. A refined fiber insulation ZOImethodology defined by WCAP-16710-P is incorporated into the IntegratedPrototype Test for chemical effects, described in the response to RAI #11 ofEnclosure 1. The methodology provides for a 7D ZOI for jacketed fiber insulationin the break selection process. Review of the break locations evaluated usingthe 17D ZOI for fiber insulation show that the original limiting break for debrisgeneration remains bounding for McGuire Units 1 and 2 when the 7D ZOI isapplied to jacketed fiber insulation.

This refinement is discussed in further detail in Section 3(b) of Enclosure 2.

3(a)(2) State whether secondary line breaks were considered in theevaluation (e.g., main steam and feedwater lines) and brieflyexplain why or why not.

McGuire Response:Secondary line breaks were not considered in the evaluation of debrisgeneration. The secondary side breaks do not introduce a type of debris to theECCS sump pool different than the primary side breaks. For debris generation,the smaller secondary side breaks inside the crane wall are bounded by theprimary side breaks, and the larger primary side breaks result in more fibrousinsulation debris and RMI debris.

Secondary side breaks were therefore not considered in the evaluation of debrisgeneration, since the primary side breaks are bounding.

3(a)(3) Discuss the basis for reaching the conclusion that the breaksize(s) and locations chosen present the greatest challenge topost-accident sump performance.

McGuire Response:As identified in the response to RAI #1 of Enclosure 1, at McGuire the limitingbreak for debris generation is the RCS Hot Leg, Loop B. This limiting break is adouble-ended guillotine break (DEGB) of a primary loop, located nearest to theECCS sump and the ECCS sump strainer. This break generates a high quantityof fiber and causes the transportation of the highest amount of fiber to thestrainer.

Page 6 of 86



3(b) Debris Generation / Zone of Influence (ZOI) (excluding coatings)The objective of the debris generation / ZOI process is to determine, foreach postulated break location: (1) the zone within which the break jetforces would be sufficient to damage materials and create debris; and (2)the amount of debris generated by the break jet forces.

General McGuire Response Note:As identified by the NRC Content Guide, the responses provided below relate todebris generation/ZOI of fiber insulation and RMI at McGuire, excluding coatings.The debris generation/ZOI information relating to coatings inside containment islocated in Section 3(h) of Enclosure 2. Debris generation information relating tolatent debris inside containment is located in Section 3(d) of Enclosure 2.

3(b)(1) Describe the methodology used to determine the ZOls forgenerating debris. Identify which debris analyses used approvedmethodology default values. For debris with ZOls not defined inthe guidance report (GR) / safety evaluation (SE), or if using otherthan default values, discuss method(s) used to determine ZOI andthe basis for each.

McGuire Response:For the initial evaluation of the generation of insulation debris, the ZOls assumedwere consistent with the default values specified in the NEI 04-07 guidance,including the SER. The methodology used in the initial GSI-191 debrisgeneration evaluation for determining the break ZOls at McGuire considered thedouble-ended guillotine break of the largest RCS piping. A spherical zone ofinfluence centered at the break location is used, consistent with NEI 04-07 (andthe companion SER) guidance. This initial evaluation provided an ECCS sumpstrainer area for the modified strainer design at McGuire. The sump strainer wassubsequently increased again in the final design to provide further strainermargin.

A refined fiber insulation ZOI for jacketed insulation, using the results of specificjet impingement testing reported in WCAP-16710-P, was utilized as an input tothe Integrated Prototype Test (IPT) for chemical effects described in detail in theresponse to RAI #11 of Enclosure 1. This is the only deviation from the SERapproved methodology for insulation debris generation. A discussion of theapplication of this WCAP can be found in the response to item 3(b)(3), below.

Page 7 of 86



3(b)(2) Provide destruction ZOls and the basis for the ZOls for eachapplicable debris constituent.

McGuire Response:There are three types of insulation debris generated within the ZOIs evaluated forMcGuire: Mirror Reflective Metal Insulation (RMI), jacketed/unjacketed Nukon®fiber insulation, and jacketed Thermal-WrapR fiber insulation. All debrisgeneration estimates within the ZOIs for these types, as reported in the NEI 04-07 guidance, are determined by jet impingement testing.

Table 3B2-1 shows the ZOI radii and destruction pressures for these threeMcGuire insulation debris types. The ZOls in this Table reflect the default valuesgiven in the NEI 04-07 guidance report. ZOls for two of the insulation types wererefined further as discussed in item 3(b)(3).

Table 3B2-1ZOI Radii for McGuire Insulation Debris Types

Debris Type Destruction Pressure ZOI Radius/Breakestcio PDiameter

(psig) (L/D)

Mirror RMI 2.4 28.6Nukon®' Insulation 6 17Thermal-Wrap' 6 17Insulation

3(b)(3) Identify if destruction testing was conducted to determine ZOls. Ifsuch testing has not been previously submitted to the NRC forreview or information, describe the test procedure and resultswith reference to the test report(s).

McGuire Response:Initial ZOls for insulation debris were determined using the NEI 04-07 guidancereport and the companion SER. These ZOls were based on NRC-evaluated jetimpingement testing as described in those documents, and were used in sizingthe modified ECCS sump strainers at McGuire. Further refinements to insulationdebris ZOIs were incorporated in the Duke IPT for chemical effects as describedin the response to RAI #11 of Enclosure 1. These refined ZOIs are also basedon specific jet impingement testing on jacketed Nukon® insulation, and areidentified in the WCAP-16710-P test report: "Jet Impingement Testing toDetermine the Zone of Influence (ZOI) of Min-K and NUKON® Insulation for WolfCreek and Callaway Nuclear Operating Plants", dated October 2007. Theevaluation within WCAP-16710-P demonstrates a refined 7D ZOI for jacketedNukon® insulation. The design and properties of jacketed Thermal-Wrap® and

Page 8 of 86



jacketed Nukono insulation are sufficiently similar such that this refined ZOI canbe applied to both.

3(b)(4) Provide the quantity of each debris type generated for each breaklocation evaluated. If more than four break locations wereevaluated, provide data only for the four most limiting locations.

McGuire Response:The four most limiting locations for debris generation are those associated withthe SG loop LBLOCA breaks (Cases 1 through 4), since these breaks producethe highest Low Density Fiberglass (LDFG) contributions. The estimatedgenerated quantities of insulation debris that follow represent the amountsgenerated using the initial 17D ZOI described in items (1), (2), and (3) above(i.e., quantities unrefined by WCAP-1 671 0-P).

The quantities of each insulation debris type generated for each of the mostlimiting break locations are given in Tables 3B4-1 through 3B4-8. Note that thevalues in parentheses indicate the fraction of the total amount for a specific sizedistribution.

Table 3B4-1Non-RMI Debris Quantities - Case 1 (LBLOCA SG Loop A)

Small Unjacketed IntactDebris Type Fines Pieces Large Pieces Blankets TotalLDFG Insulation - 249.0 ft3 729.5 ft3 520.3 ft3 557.6 ftCrossover (12.4%) (35.5%) (25.3%) (27.1%) 2056.4 ft3

LDFG Insulation - 262.7 ft3 840.0 ft3 445.2 ft3 476.9 ft3

Hot Leg (13.0%) (41.5%) (22.0%) 0 (23.5%) 2024.8 ft3

Table 3B4-2RMI Debris Quantities - Case 1 (LBLOCA SG Loop A)

Debris Type Amount Destroyed by Size DistributionTotal AmountDestroyed 1/4" " 1" 2" 4" 6"

1,372 ft2 6,443 ft2 6,667 ff2 8,166 ft2 5,359 ff2 3,892 ft2

31,898.2 ft (4.3%) (20.2%) (20.9%) (25.6%) (16.8%) (12.2%)

Table 3B4-3Non-ReI Debris Quantities - Case 2 (LBLOCA SG Loop 1)

Small Unjacketed IntactDebris Type Fines Pieces Large Pieces Blankets TotalLDFG Insulation - 256.4 ft3 766.1 ft3 515.5 ft3 552.5 f 3

Crossover (12.3%) (36.6%) (24.7%) (26.4%) 2090.5 ft3


Hot Leg (13.1%) (42.8%) (21.3%) (22.8%) 2085.9 ft3

Page 9 of 86



Table 3B4-4RMI Debris Quantities - Case 2 (LBLOCA SG Loop B)

Debris Type Amount Destroyed by Size DistributionTotal AmountDestroyed 1/4 1/2" 1" 2" 4" 6"

1,347 ft2 6,329 ft2 6,549 ft2 8,021 ft2 5,264 ft2 3,823 ft2

31,334 ft2 (4.3%) (20.2%) (20.9%) (25.6%) (16.8%) (12.2%)

Table 3B4-5Non-RMI Debris Quantities - Case 3 (LBLOCA SG Loop C)

Small Unjacketed IntactDebris Type Fines Pieces Large Pieces Blankets TotalLDFG Insulation - 232.0 ft3 711.6 ft3 438.2 ft3 469.5 ft3

Crossover (12.5%) (38.4%) (23.7%) (25.4%) 1851.3 f3


Hot Leg (13.8%) (47.5%) (18.7%) (20.0%) 1770.5 ft3

Table 3B4-6RMI Debris Quantities - Case 3 (LBLOCA SG Loop C)

Debris Type Amount Destroyed by Size DistributionTotal AmountDestroyed 114" 1/2" 1 2" 4" 6"

1,188 ft2 5,579 ft2 5,773 ft2 7,071 ft2 4,640 ft2 3,370 ft227,621 ft2 4.3% 20.2% 20.9% 25.6% 16.8% 12.2%

Table 3B4-7Non-RMI Debris Quantities- Case 4 (LBLOCA SG Loop D)

Small Unjacketed IntactDebris Type Fines Pieces Large Pieces Blankets TotalLDFG Insulation - 226.8 ft3 691.8 ft3 435.6 ft3 466.8 ft3

Crossover (12.5%) (38.0%) (23.9%) 25.6% 1821.0 ft3

LDFG Insulation - 242.3 ft3 815.8 ft3 355.3 ft3 380.4 ftHot Leg (13.5%) (45.5%) (19.8%) (21.2%) 1793.8 ft3

Table 3B4-8RMI Debris Quantities- Case 4 (LBLOCA SG Loop D)

Debris Type Amount Destroyed by Size DistributionTotal AmountDestroyed ¼" 12" 1" 2" 4" 6"

1,238 ft2 5,818 ft2 6,020 ft2 7,373 ft2 4,839 ft2 3,514 ft2

28,802 ft2 (4.3%) (20.2%) (20.9%) (25.6%) (16.8%) (12.2%)

3(b)(5) Provide total surface area of all signs, placards, tags, tape, andsimilar miscellaneous materials in containment.

McGuire Response:Signs, placards, tags, tape, and similar miscellaneous materials in containment(including dust, dirt, and lint) are defined, for the purposes of the McGuire

Page 10 of 86



modified ECCS sump strainer design and debris generation evaluation, as latentdebris and miscellaneous latent debris. The total quantity of latent debris insidecontainment, and the amount of latent debris estimated to be generated as aresult of a LBLOCA, is discussed in detail in Section 3(d) of Enclosure 2.

Page 11 of 86



3(c) Debris CharacteristicsThe objective of the debris characteristics determination process is toestablish a conservative debris characteristics profile for use indetermining the transportability of debris and its contribution to head loss.

3(c)(1) Provide the assumed size distribution for each type of debris.

McGuire Response:Three types of potentially transportable debris are generated at McGuire during apostulated accident: destroyed insulation (fibrous and RMI), failed coatings, andlatent debris.

As described in Enclosure 2, Section 3(b), for the initial evaluation of thegeneration of insulation debris, the ZOls assumed are consistent with the defaultvalues specified in the NEI 04-07 guidance, including the SER.

A refined fiber insulation ZOI and size distribution for jacketed insulation, usingthe results of specific jet impingement testing reported in WCAP-16710-P, wasutilized as an input to the Integrated Prototype Test (IPT) for chemical effectsdescribed in detail in the response to RAI #11 of Enclosure 1. This is the onlydeviation from the SER approved methodology for insulation debris generation. Adiscussion of the application of this WCAP can be found in Enclosure 2, Section3(b). The results of the IPT employing these refinements are addressed in theresponse to RAI #12 of Enclosure 1.

Table 3C1 -1 below shows the results of an analysis that supports the use ofspecific size distributions for the initial fiber insulation debris at McGuire inComputational Fluid Dynamics (CFD) analyses. This analysis utilizes guidancefound in the NEI-04-07 SER. For Nukon® and Thermal Wrap® insulation, it wasdetermined that overall fibrous debris size distribution is best defined using threeZOI sub-zones.

Page 12 of 86



Table 3C1-1

McGuire Initial LDFG Debris Distributions

18.6 psi ZOI 10.0-18.6 psi ZOI 6.0-1!0.0 psi ZO!SIZE (7.0 L/D) (11.9-7.0 L/D) (17.0-11.9 LiD)

Fines (individual Fibers) 20% 13% 8%

Small Pieces (<6" on a side) 80% 54% 7%

Large Pieces (>6" on a side) 0% 16% 41%

Intact (covered) Blankets 0% 17% 44%

The size distribution of destroyed RMI, primarily Diamond Power Mirror Insulationat McGuire, is depicted in Figure 3C1-1 below. The destruction pressure for thistype of insulation with standard banding is 2.4 psi, which corresponds to a ZOIradius of 28.6 pipe diameters as identified by the NEI-04-07 SER. The sizedistribution for RMI, as provided in NEI-04-07 guidance and the companion SER,is 75% small pieces and 25% large pieces.

30%

0.I

20%

15%

10%

5%

0%

1/4" 1V2" 1. 2"

Debris Size

a"

Figure 3C1-1McGuire RMI Debris Size Distribution

Page 13 of 86



Debris sizes for failed coatings and latent debris (dust, dirt, and lint) can be foundin the response to item 2 below. Miscellaneous latent debris (stickers, labels,and tags) is assumed to have various sizes, and is addressed in Section 3(d) ofEnclosure 2.

3(c)(2) Provide bulk densities (i.e., including voids between thefibers/particles) and material densities (i.e., the density of themicroscopic fibers/particles themselves) for fibrous andparticulate debris.

McGuire Response:NEI 04-07 and the companion SER (Method 2) provide a conservative estimateof the densities of the latent fibers and particulates (94 lb/ft3 and 169 Ib/ft3,respectively). To be consistent with the head loss analysis, the microscopicdensity of the latent fiber material. is conservatively assumed equivalent to that ofNukon® fiberglass (175 Ib/ft3). The NRC SER also states that the particulate sizecan be estimated by using the NUREG/CR-6224 head loss data for typicalmixtures of latent particulate debris. The latent particulate debris size using thismethodology is 17.3 microns. Additionally, the NRC SER (Method 2) states thatthe latent fiber sizing for head loss purposes is assumed to be the same asreported in NUREG/CR-6224 for commercial fiberglass (approximately 7microns).

The densities of the different debris types generated at McGuire are summarizedin Table 3C2-1 below:

Table 3C2-1McGuire Generated Debris Characteristics

Debris Material Macroscopic Microscopic Characteristic CharacteristicDensity Density Size Size(lb/ft3) (Ib/ft3) (pm) (ft)

Fiberglass Insulation 2.4 175 7.112* 2.33E-05

Latent Fibers 2.4 175*** 7.112* 2.33E-05

Qualified Coatings - N/A 118 10"* 3.28E-05EpoxiesUnqualified Coatings N/A 94 & 98 10** 3.28E-05

Latent Dirt/Dust N/A 169 17.3** 5.68E-05* fiber diameter

** - spherical particle diameter- latent fiber microscopic density of Nukon® insulation to be consistent with head loss analysis

Page 14 of 86



3(c)(3) Provide assumed specific surface areas for fibrous andparticulate debris.

McGuire Response:The specific debris characteristic sizes used for fibrous and particulate debris inthe head loss analysis are provided in Table 3C2-1 in the response to item 2,above.

3(c)(4) Provide the technical basis for any debris characterizationassumptions that deviate from NRC-approved guidance.

McGuire Response:Initial debris characterization assumptions used in the evaluation of estimatedgenerated debris at McGuire are primarily taken from the guidance provided byNEI 04-07 and the associated NRC SER. For the determination of the McGuirefibrous insulation debris size distribution, an extension of the methodology andguidance provided in the NEI 04-07 SER was necessary.

For a baseline analysis, the NEI-04-07 guidance document recommends a sizedistribution with two categories: 60% small fines and 40% large pieces. The SER(Appendix VI, Section 3.2) suggested a more refined approach for determiningthe debris size distribution based on applicable air jet impact tests. UsingAppendices II and VI from the SER, a debris size distribution for Nukon® (and viasimilarity, Thermal-Wrap®) insulation was developed. It was determined thatwithin the overall break ZOI, the size distribution of fibrous insulation would varybased on the distance of the insulation from the break (i.e., insulation debrisgenerated near the break location would consist of more small pieces thaninsulation debris generated near the edge of the ZOI). The response to item3(c)(1) above provides specifics regarding the results of this extended analysisand the resulting assumed size distributions.

Other assumptions regarding debris characteristics are located in the responseto RAI #1 of Enclosure 1.

Further refinements (i.e., to jacketed fiber insulation ZOIs and to assumptionsrelated to failed unqualified epoxy coatings) utilizing specific industry testing areincorporated in the Integrated Prototype Test for chemical effects as describedby RAI #11 of Enclosure 1, and in Sections 3(b) and 3(h) of Enclosure 2.

Page 15 of 86



3(d) Latent DebrisThe objective of the latent debris evaluation process is to provide areasonable approximation of the amount and types of latent debris existingwithin the containment and its potential impact on sump screen head loss.

3(d)(1) Provide the methodology used to estimate quantity andcomposition of latent debris.

McGuire Response:Latent debris, for the purposes of the modified McGuire ECCS sump strainerdesign and evaluation, is defined as dirt, dust, paint chips, fibers, paper scraps,plastic tags, tape, adhesive, labels, fines or shards of thermal insulation, fireproofbarrier, "owner-installed" material (e.g. signs, stickers, etc.), or other materialsthat may be present in containment prior to a postulated LOCA.

As discussed in the response to RAI #34 of Enclosure 1, McGuire containmentforeign materials walkdowns were conducted using NEI 02-01 guidance for bothUnits. As a part of these walkdowns, the existence of latent debris wasevaluated. The walkdown results were tabulated using walkdown notes andphotographs. Only materials that were expected to remain in containment afteran outage were included in the inventories.

Subsequent to these walkdowns, a tag and label reduction evaluation wasperformed to analytically reduce the amount of stickers, labels, and tags thatcould fail in a postulated LOCA and transport to the ECCS sump pool, usingcurrent EQ qualifications and engineering judgment.

An additional 20% was then added to take into account missed materials, areasof low photograph-to-area size ratios, and inaccessible areas due to limitedspace and high radiation.

The latent debris tabulations were used to develop a reasonable but conservativetotal square footage of each material by containment area. Generic samplingdata (mass densities) from other plants, combined with subjective walkdownobservations as to plant cleanliness, were also used to make quantitativeestimates of latent debris by containment area.

Page 16 of 86



3(d)(2) Provide the basis for assumptions used in the evaluation.

McGuire Response:The following discussion provides the assumptions and their bases regarding thetreatment of latent debris inside the McGuire containments:

" The amount of dust, dirt, and lint was estimated to be about 90 lb in Unit 2,and 140 lb in Unit 1. A bounding value of 200 lb was used in the debrisgeneration evaluation to provide adequate margin.

* Penetration sealant is assumed to fail only in the break ZOI. Foam sealantsare identified only in the upper containment. There are no breaks postulatedin upper containment.

" The walkdown report identifies flexible connections in various areas incontainment as miscellaneous debris. It is assumed that only the flexibleconnections within the break ZOI will be destroyed. The only flexibleconnections identified in the walkdown report are in the Ice Condenser. Thereare no breaks postulated in the Ice Condenser area. It is reasonable toassume that flexible connections that are outside a break ZOI will notspontaneously fail.

* Per NEI 04-07, the fiber content of the latent dust and dirt debris is assumedto be 15% by mass. With the assumption of 200 lb of latent debris, 30 lb ofthe debris is considered to be latent fibers. The NRC SER for NEI 04-07further assumes that the latent fiber bulk density is assumed to be the sameas low density fiberglass material (2.4 lb/ft3). This results in 12.5 ft3 of latentfibrous debris. NEI 04-07 and the NRC SER Method 2 provide a conservativeestimate of the latent fibers and particulate densities (94 Ib/ft 3 and 169 lb/ft3,respectively). To be consistent for the McGuire head loss analysis, themicroscopic density of the latent fiber material was assumed to be equivalentto Nukon® fiberglass (175 lb/ft3). The NRC SER also states that theparticulate size can be estimated by using the NUREG/CR-6224 head lossdata for typical mixtures of latent particulate debris. The latent particulatedebris size using this methodology is 17.3 microns. Additionally, the SERstates that the latent fiber sizing for head loss purposes are assumed to bethe same as reported in NUREG/CR-6224 for commercial fiberglass(approximately 7 microns).

The following discussion provides the assumptions and their bases for the Tagand Label reduction evaluation in the McGuire containment buildings:

* All assumed percentages are estimated from plant drawings, walkdownexperience, and walkdown photos. All percentages were initially estimated

Page 17 of 86



and then adjusted to provide conservatism; however, all are based onengineering judgment. The reductions will be applied to the actual tag or labelcounts and then rounded up to the nearest whole number (i.e. there are nopartial tags or labels).

" All tags and labels that detach from their affixed positions are assumed to fallstraight down when in the presence of containment spray only (i.e. nosubmergence and no jet impingement). The same assumption applies whencontainment spray is not present, such as in the accumulator and fan roomsabove the maximum flood level.

" A large portion of the tags and labels inside the crane wall in lowercontainment will be in the break's zone of influence (ZOI) and will fail. It is notpossible to conservatively estimate the percentage of tag and label surfacearea that is in the ZOI; therefore, all tags and labels inside the crane wall inlower containment will be assumed to fail.

" Plastic tags outside the ZOI are assumed to stay intact. While there may besome deformation due to the LOCA environment, they are assumed to notbecome overly pliable (i.e. they will not deform to pass through an obstructionthat has a smaller dimension than the tag).

3(d)(3) Provide results of the latent debris evaluation, including amountof latent debris types and physical data for latent debris asrequested for other debris under c. above.

McGuire Response:Latent debris quantities are summarized in Tables 3D3-1 and 3D3-2 below.

Table 3D3-2 represents the McGuire Unit 2 latent debris quantities which wereassumed bounding for Unit 1; this assumption was subsequently verified. Thequantities tabulated include a tag and label refinement evaluation performedutilizing current EQ qualifications and engineering judgment, as identified in theresponse to item 3(d)(1) above.

Table 3D3-1McGuire Latent Debris Quantity (Dust, Dirt, and Lint)

Latent Debris Type Weight Volume

Dirt and Dust 170 lb N/A

Latent Fibers (lint) 30 lb 12.5 ft3

Based on a bulk density of 2.4 lb/ft3, similar to LDFG (see assumptions in response to item 2)

Page 18 of 86



Physical data for dust, dirt, and lint debris types can be found in the response toitem 3(d)(2) above.

Table 3D3-2McGuire Refined Miscellaneous Latent Debris Quantities

Type of Debris Lower Containment Lower Containment Upper Ice Total

(Inside Crane Wall) (Outside Crane Wall) Containment Condenser

Stickers & Labels

(ft2) 64.96 30.34 14.43 8.92 118.65Plastic Tags w/Adhesive (ft2) 33.32 8.79 11.23 8.02 61.36Plastic HangingTags (ft2) 11.70 4.33 N/A 0.17 16.20RMI ID Stickers

(ft2) 103.14 32.81 N/A N/A 135.95Ice Condenser

Debris (ft2) N/A N/A N/A 15.30 15.30

Total (ft2) N/A N/A N/A N/A 347.5

3(d)(4) Provide amount of sacrificial strainer surface areamiscellaneous latent debris.

allotted to

McGuire Response:The miscellaneous latent debris total area contribution to sump strainer blockageat McGuire is 347.5 square feet, as shown in Table 3D3-2 in the response to item3(d)(3) above.

NEI 04-07 guidance recommends that 75% of the total miscellaneous latentdebris transporting to the ECCS sump pool be allotted to sump strainer blockage.

Page 19 of 86



3(e) Debris TransportThe objective of the debris transport evaluation process is to estimate thefraction of debris that would be transported from debris sources withincontainment to the sump suction strainers.

3(e)(1) Describe the methodology used to analyze debris transportduring the blowdown, washdown, pool fill-up, and recirculationphases of an accident.

McGuire Response:The methodology used to analyze debris transport during the blowdown,washdown, pool fill-up, and recirculation phases of an accident is as follows:

1. Based on containment building drawings, a three-dimensional model wasbuilt using computer aided drafting (CAD) software.

2. A review was made of the drawings and CAD model to determine transportflow paths. Potential upstream blockage points including screens, grating,drains, etc. that could lead to water holdup were addressed.

3. Debris types and size distributions were gathered from the debris generationcalculation for each postulated break location.

4. The fraction of debris blown to various areas of containment was determinedbased on the flow of steam during the blowdown.

5. The quantity of debris washed down by ice melt and spray flow wasconservatively determined.

6. The quantity of debris transported to inactive areas or directly to the sumpscreen during pool fill-up was determined to be negligible.

7.- Using conservative assumptions, the locations of each type/size of debris atthe beginning of recirculation was determined.

8. A Computational Fluid Dynamics (CFD) model was developed to simulatethe flow patterns that would occur during recirculation. Further detailsregarding the CFD model are located in the response to item 3 below.

9. A graphical determination of the transport fraction of each type of debris wasmade using the velocity and Turbulent Kinetic Energy (TKE) profiles fromthe CFD model output, along with the determined initial distribution of debris.

10. The recirculation transport fractions from the CFD analysis were gathered toinput into the logic trees.

11. The quantity of debris that could experience erosion due to the break flow,spray flow, or ice melt drainage was determined.

Page 20 of 86



The overall transport fraction for each type of debris was determined bycombining each of the previous steps in the logic tree.

3(e)(2) Provide the technical basis for assumptions and methods used inthe analysis that deviate from the approved guidance.

McGuire Response:The methodology used in the McGuire debris transport analysis is based on theNEI 04-07 guidance report for refined analyses, as modified by the NRC SER, aswell as the refined methodologies suggested by the SER in Appendices III, IV,and VI. The specific effect of each mode of transport was analyzed for each typeof debris generated, and a logic tree was developed to determine the totaltransport to the ECCS sump strainer. Assumptions used in the McGuire debristransport analysis are listed in the response to RAI #1 of Enclosure 1.

The following methodology used in the evaluation deviates from the NEI 04-07guidance and the companion NRC SER:

The logic tree approach was different than the baseline logic tree provided inthe NEI 04-07 guidance report. The change was made to account for non-conservative assumptions identified by the NEI 04-07 SER, including thetransport of large pieces, erosion of small and large pieces, the potential forwashdown debris to enter the pool after inactive areas have been filled, andthe direct transport of debris to the sump strainer during pool fill-up.

NEI 04-07 Section 3.4.3 recommends using a two-category size distributionfor insulation debris including: (1) small pieces (assumed to be the basicconstituent of the material), and (2) large pieces (pieces greater than 4inches). Although adequate, this size distribution allows for only limitedbenefit when CFD analyses are used to refine the recirculation pool debristransport fractions. The NRC recognized this limitation in their NEI 04-07SER. SER Section 4.2.4 recommends a four-category size distribution:

" Fines that remain suspended* Small piece debris that is transported along the pool floor* Large piece debris with the insulation exposed to potential erosion* Large debris with the insulation still protected by a covering, thereby

preventing erosion

The methodology that can be used to determine the fraction of debris fallingwithin each of the four categories was explained in Appendices II and VI ofthe SER, but was not fully carried out.

Page 21 of 86



For McGuire debris transport analysis, a complete methodology necessaryfor assigning a four-category size distribution for low density fiberglass(LDFG) was utilized for initial debris generation calculations. Analysis ofLDFG insulating materials demonstrates that the fraction of fines and smallpieces decreases with increasing distance from the break jet, and thefraction of large pieces and intact blankets increases with increasingdistance for LDFG.

Additionally, a refined ZOI and size distribution for jacketed fiber insulation,using the results of specific jet impingement testing reported in WCAP-16710-P, was utilized as an input to the Integrated Prototype Test (IPT) forchemical effects described in detail in the response to RAI #11 of Enclosure1. A discussion of the application of this WCAP can be found in Enclosure2, Section 3(b). The results of the IPT employing these and otherrefinements are addressed in the response to RAI #12 of Enclosure 1.

3(e)(3) Identify any computational fluid dynamics codes used to computedebris transport fractions during recirculation and summarize themethodology, modeling assumptions, and results.

McGuire Response:The CFD calculation for recirculation flow in the McGuire containment sump poolwas performed using Flow-3D® Version 9.0.

The CFD model was generated based on the following characteristics:

1. The mesh in the CFD model was nodalized to sufficiently resolve thefeatures of the CAD model, but still keep the cell count low enough for thesimulation to run in a reasonable amount of time.

2. The boundary conditions for the CFD model were set based on theconfiguration of McGuire during the recirculation phase.

3. The ice melt and containment spray flows were included in the CFDcalculation, with the appropriate flowrate and kinetic energy to accuratelymodel the effects on the containment sump pool.

4. At the postulated break location, a mass source was added to the model tointroduce the appropriate flowrate and kinetic energies associated with thebreak flow.

5. A negative mass source was added at the sump pool location with a totalflowrate equal to the sum of the spray flow and break flow.

6. An appropriate turbulence model was selected for the CFD calculations.

Page 22 of 86



7. After running the CFD calculations, the mean kinetic energy was checked toverify that the model had been run long enough to reach steady-stateconditions.

8. Transport metrics were determined based on relevant tests and calculationsfor each significant debris type present in the McGuire containment breakZOls.

Assumptions used in the CFD model are as follows:

1 . It was assumed the recirculation transport fractions determined for theMcGuire Loop B break can be applied to the other breaks inside the cranewall. This is a conservative assumption since the Loop B break location isclosest to the ECCS sump strainer.

2. The water falling from the reactor coolant system break was assumed to doso without encountering any structures before reaching the containmentsump pool. This is a conservative assumption since any impact withstructures would dissipate the momentum of the water and decrease theturbulent energy in the pool.

3. It was assumed that potential upstream blockage points (e.g., drains,fences, grating, etc.) would not inhibit the flow of water.

Logic trees were used to determine the fractions of the various types of debristhat would reach the containment sump pool. Since the recirculationtransport fractions are assumed to be the same for each of the breaks insidethe crane wall, the overall transport fraction would also be the same. Logictrees were constructed for small RMI debris, large RMI debris, small lowdensity fiberglass debris, and large low density fiberglass debris. For all RMIdebris, it was determined that no debris would transport to the active pool. Inaddition, since the latent fiber, dirt/dust, and paint particulate were allassumed to reach the recirculation pool and the recirculation transportfraction is 100%, the overall transport fraction is also 100%.

Logic trees are shown in Figures 3E3-1 and 3E3-2 below for small piecefiberglass debris and large piece fiberglass debris, respectively:

Page 23 of 86


2004-02 Supplemental ResponsesMcGuire Nuclear Station.

Blowdown Washdown Pool Fill CFD Fraction ofDebris Size Transport Transport Transport Recirculation Erosn Debris at Sump

Transport

0.03001Active Pool 0.10 0.002

0.66 Erodes to Fines

Sediment 0.900,03 Remains Intact

Pipe ChaseSump Screens

0.00

Small Inactive Pool

FiberSlass 0.12 0.116Debris Transport

Generation 1000Active Pool 0.10 0.005

Fn Erodesato Fines

Sediment 0.900.97 Remains Intact-

Inside CraneWall 0.00 .000

Sump Screens

Inactive Pool

Sum: 0.2100

Figure 3E3-1

McGuire Small Piece Fiberglass Debris Transport Logic Tree

Debris Size Blowdown [ Washdown PooI Filil CFD ý raction ofTs 10Recirculation Erosion I erstSm

Transport Tanspr Transpor TranslportD eriaum

0.00 0.000•Transport

t .00Active Pool 0.10 0.001

1.00 |Erodes to FinesSediment ] 0.90

0.01 Remains IntactPipe Chase 01 .0

Sump Screens

0.00Large Inactive Pool

Fiberglass 0.00 0.000Debris Transport

Generation 1.00Actlive Pool 0.10 0.099

I no Erodes to Fines

Sediment 0.90

0.99 Remains IntactInside Crane

Wall 0.00 0.000Sump Screens

0.00Inactive Pool

Sum: 0,100

Figure 3E3-2

McGuire Large Piece Fiberglass Debris Transport Logic Tree

Page 24 of 86



3(e)(4) Provide a summary of, and supporting basis for, any credit takenfor debris interceptors.

McGuire Response:The design and placement of the McGuire modified ECCS sump strainerprovides for the filtration of large debris entrained in the sump pool prior toreaching the strainer via passage of water through openings in the crane wall(credited) for those portions of the strainer located outside the crane wall, orthrough 3/32 inch openings in the strainer enclosure (not credited) for that portionof the strainer located inside the crane wall.

For the credited flowpath, water reaches the sump strainer after first passingthrough the crane wall penetrations. Thus, entrained debris is allowed to settleand be filtered by passage through crane wall penetrations prior to reaching thestrainer. For the uncredited flowpath, water reaches the strainer by passingthrough the ECCS strainer enclosure. The enclosure surrounding the strainerportion located inside the crane wall has perforated sides with 3/32 inch diameterholes. This path is not credited due to the potential for blockage by the projectedfiber debris load. Either flowpath will effectively prevent large debris fromreaching the strainer assemblies.

3(e)(5) State whether fine debris was assumed to settle and providebasis for any settling credited.

McGuire Response:As discussed in the response to RAI #37 of Enclosure 1, for each postulatedbreak, fine debris (i.e., dust, dirt, lint, and failed coatings particulates) wasassumed to transport 100% to the McGuire containment sump pool. Upstreamfine debris settling is not credited in the head loss testing nor in the analyticaldesign basis of the McGuire modified ECCS sump strainers.

3(e)(6) Provide the calculated debris transport fractions and the totalquantities of each type of debris transported to the strainers.

McGuire Response:Of the five McGuire postulated breaks, the most limiting break is the ReactorCoolant System Hot Leg break in Steam Generator Loop B. The initial calculateddebris transport fractions and the total quantities of each type of debristransported to the strainers for the Case 2 Hot Leg break are given in Table 3E6-1 below. As discussed in item 3(e)(2) above, refinements to the ZOI and size

Page 25 of 86



distribution for jacketed fibrous insulation are being incorporated into the finalECCS sump strainer performance evaluation.

Table 3E6-1

Initial Debris Transport to McGuire ECCS Sump StrainersCase 2 (LBLOCA SG Loop B Hot Leg)

Debris Debris DebrisQuantity Transport Quantity at

Debris Type Debris Size Generated Fraction Sump

Small Pieces (<4") 22,246 ft2 0% 0 ft2

Stainless Steel RMI Large Pieces (>4") 9,087 ft2 0% 0 ft2

Total 31,333 ft2 0% 0 ft2

Fines 272.7 ft3 100% 272.7 ft3

Nukon® and Thermal- Small Pieces (<6") 891.9 ft3 21% 187.3 ft3

Wrap® LDFG (Hot Leg Large Pieces (>6") 444.9 ft 3 10% 44.5 ft3

Break) Intact Pieces (>6") 476.4 ft3 0% 0 ft3

Total 2,085.9 ft3 24% 504.5 ft3

Qualified Epoxy (5D ZOI) Total (fines) 167.6 lb 100% 167.6 lb

Unqualified Epoxy Total (fines) 654.2 lb 100% 654.2 lb

Unqualified Alkyd Total (fines) 15.7 lb 100% 15.7 lb

Dirt/Dust Total (fines) 170 lb 100% 170 lb

Latent Fiber Total (fines) 12.5 ft3 100% 12.5 ft3

Other Latent Debris Total 347.5 ft2 100% 347.5 ft2

Page 26 of 86



3(f) Head Loss and Vortexing

The objectives of the head loss and vortexing evaluations are to calculatehead loss across the sump strainer and to evaluate the susceptibility of thestrainer to vortex formation.

General McGuire Response NoteThe head loss testing and analysis described in this Section reflect the McGuireECCS sump strainer in a clean and debris loaded condition (i.e., fiber,particulate, and latent debris). Chemical effects on the debris loaded condition ofthe strainer are addressed in Section 3(6) of Enclosure 2.

3(f)(1) Provide a schematic diagram of the emergency core coolingsystem (ECCS) and containment spray systems (CSS).

McGuire Response:The ECCS schematic for McGuire is shown in Figure 3F1-1 following. The CSSschematic for McGuire is shown in Figure 3F1-2 following.

Page 27 of 86



Oj.

K & 0!. PEw ntI7 t2

n A..r.A,

Jm ;A IC Me

U)

1 m m 00 m o

25" 895 X

50

F.-

Figure 3F1-1

McGuire Emergency Core Cooling System

Page 28 of 86



Auxiliary Spray ~ RHR

25

Auxiliary Spray . ."I .. I , RHRL& 38 B

20

LO

32 18

_ A

•29 RHR NI8

McurLO CnannSpyContainmeSf P tm1i sa Sump

S15•

12

Figure 3F11-2

McGuire Containment Spray System

3(f)l(2) Provide the minimum submergence of the strainer under small-

break loss-of-coolant accident (SBLOCA) and large-break loss-of-coolant accident (LBLOCA) conditions.

McGuire Response:As discussed in the response to RAI #41 of Enclosure 1, the limiting analyticalcase for minimum ECCS sump level. at McGuire is characterized as a smallbreak LOCA (SBLOCA) during which Containment Spray does not actuate andthere is no water source contribution from ice melt. The minimum submergence

Page 29 of 86



of the McGuire ECCS sump strainer under postulated SBLOCA conditions is atleast 2 inches.

The various large break LOCA (LBLOCA) cases generate more water and moresubmergence than the limiting SBLOCA case outlined above. The LBLOCAcases will have an Ice Condenser contribution to ECCS sump inventory due toice melt, -and at larger postulated break sizes, additional containment sump poolcontributions from the RCS and the Cold Leg Accumulator Tanks.

3(f)(3) Provide a summary of the methodology, assumptions and resultsof the vortexing evaluation. Provide bases for key assumptions.

McGuire Response:NUREG-0897 summarizes the results of testing performed at Alden ResearchLaboratory that determined the susceptibility of forming a vortex in typical PWRsumps. The results of this testing are also contained within Regulatory Guide1.82. This testing was based on sump layouts that included an open pit withsingle or dual horizontal and vertical intakes and a screen outside and above thepit. For McGuire, the entire 725'+0" elevation is considered the "sump" and thereis no actual pit. The new strainer (including top-hats, flow plenums, andwaterboxes) is located on the floor in the sump with multiple suction points intothe strainer, and a much greater strainer surface area from which to draw flow.Therefore, the vortex formation parameters presented in Regulatory Guide 1.82are considered to be overly conservative.

The top-hat strainer modules are completely covered by horizontal standard floorgrating for the purpose of vortex suppression. The top of the grating inside thecrane wall is also covered by 14 gauge solid plate. The minimum containmentwater level is at least 3 feet above the sump floor; at this water level, both vortexsuppression gratings are fully submerged (by at least 2 inches as notedpreviously) and provide assurance that the suction lines will not be susceptible toair ingestion caused by air core vortex formation from the post-LOCAcontainment building water surface.

Top-hat strainer module testing demonstrates that standard floor gratingeliminates air core vortices for top-hat approach velocities ranging from 0.01ft/sec to 0.09 ft/sec. This testing was performed with a few inches of watercoverage above the top hat modules similar to the top hat modules at McGuire.The maximum approach velocity for the top hat modules is approximately 0.052ft/sec. Since the maximum approach velocity for the top-hat modules at McGuireis within the tested flow condition, and since the submergence is consistent withthe tested condition, the McGuire top-hat strainer modules are not susceptible toair ingestion from an air core vortex.

Page 30 of 86



The vortexing evaluation is described further and summarized in the response toRAI #39 of Enclosure 1.

3(f)(4) Provide a summary of the methodology, assumptions, and resultsof prototypical head loss testing for the strainer, includingchemical effects. Provide bases for key assumptions.

McGuire Response:A detailed discussion of the methodology, design inputs, and assumptionsregarding Duke's prototypical head loss testing for the McGuire ECCS sumpstrainer top-hat modules (i.e.,, the Integrated Prototype Test (IPT) includingchemical effects) is located in the response to RAI #11 of Enclosure 1.

Initial prototypical head loss testing for the McGuire top-hat strainer modules wasperformed during an array test, during which various quantities of debris (i.e.,fiber and particulate) were allowed to collect on the modules while head lossmeasurements were made. The data collected from this testing was used togenerate a head loss correlation which is used in determining the total head lossacross the ECCS sump strainer. Chemical effects on the strainer debris bed aredemonstrated via a different prototype test as identified below.

The total calculated head loss of the strainer consists of four parts:

" Head loss across the debris bed (based on top-hat module array test data)

" Head loss across the clean perforated plate mesh surfaces in the top-hats(based on hydraulic analysis)

* Head loss through the waterbox/plenum arrangement connecting the arrayof top-hats to the ECCS suction piping (based on hydraulic analysis)

" Head loss due to cumulative chemical effects across debris-loaded top-hats(based on the IPT chemical effects data)

The response to RAI #12 of Enclosure 1 indicates that the final head losscalculation will be performed and documented upon the finalization of the IPT.Upon completion of the IPT documentation, a refined debris load head losscalculation will be generated that incorporates any added consequence of testedchemical effects. This is a commitment identified in Duke letter dated November6, 2007, "Request for Extension of Completion Dates for McGuire Units 1 and 2Corrective Actions Required by NRC Generic Letter (GL) 2004-02". Dukeexpects to have this documentation, and to supplement the response to RAI #12,by April 30, 2008.

Page 31 of 86

Enclosure 2Information Addressing Issues Identified in Staff Content Guide.for Generic Letter


3(f)(5) Address the ability of the design to accommodate the maximumvolume of debris that is predicted to arrive at the screen.

McGuire Response:As discussed in the response to RAI #40 of Enclosure 1, the predicted head lossfor the McGuire modified ECCS sump strainer utilizes test data obtained from aprototypical top-hat module array (2 high x 3 wide). The top-hats used in thisarray testing were 36 inches long, as this length is representative of thepopulation of top-hat modules on the McGuire strainers. The array test useddebris loads (i.e., particulates and fiber) in various quantities postulated totransport to the containment sump after a large-break LOCA. Section 3(e) ofEnclosure 2 discusses the debris quantities expected to transport to the McGuireECCS sump strainer.

Expected Design BehaviorThe debris bed initially accumulates non-uniformly on the top-hat. The approachvelocity will vary across the individual top-hats and across the array based on thelocation of the top-hats relative to the suction source. As the debris bed buildsup to the maximum load, the debris bed starts to fill the interstitial volume andbegins to transition to the circumscribed area of the strainer. Transitioning to thecircumscribed area changes the strainer from a complex shape (multiplecylinders with flow passages outside and inside the cylinder) to a simplecylindrical shape with a single outer flow passage. This transition results in adecreased surface area and increased head loss. The debris bed at this point isalso more uniform than the thinner beds and results in increased head loss. Asthe debris bed is more uniform for the maximum load, flow through the debrisbed is more uniform and head loss is governed by the bed thickness andapproach velocity.

3(f)(6) Address the ability of the screen to resist the formation of a "thinbed" or to accommodate partial thin bed formation.

McGuire Response:The thin-bed effect is defined as the relatively high head losses that occur acrossa uniform thin bed of fibrous debris that can sufficiently filter particulate debris toform a dense particulate debris bed. The thin-bed effect is typically seen intesting of a strainer with a simple geometry such as a flat plate. Strainer designswith a more complex geometry are more likely to load non-uniformly, precludingthe formation of a thin bed.

The top-hat modules used on the modified McGuire ECCS sump strainersconsist of hollow concentric cylinders mounted on a square base. The cylindersare comprised of stainless steel perforated plate.

Page 32 of 86



A series of tests were performed on this top-hat module design for the purpose ofdetermining the head loss at high particulate/fiber ratios. The measured headloss conservatively bounded any head loss that could be achieved with varyingfiber quantities. No indication of a thin-bed effect was observed.

Based on results of this testing, it can be concluded that the modified McGuireECCS sump strainer utilizing an array of strainer modules (top-hats) that do notexhibit thin-bed formation.

The Integrated Prototype Test (IPT) was designed to replicate this feature of thetop-hat modules also. The results of this testing, to be reported in spring 2008 asnoted in item 3(f)(4) above, are expected to further confirm this conclusion.

3(f)(7) Provide the basis for the strainer design maximum head loss.

McGuire Response:The predicted maximum head loss across the strainer is associated with themaximum debris generation case, the maximum debris transport to the ECCSsump pool, the maximum flowrates in the ECCS sump pool, and the lowest sumppool temperature.

3(f)(8) Describe significant margins and conservatisms used in the headloss and vortexing calculations.

McGuire Response:The.assumptions and conservatisms included in the McGuire debris generationevaluation, the debris transport evaluation, and the vortex suppression evaluationare listed and discussed in the response to RAIs #1 and #39 of Enclosure 1. Inaddition, Sections 3(b), 3(c), 3(d), and 3(h) of Enclosure 2 detail manyconservatisms incorporated into the postulated debris challenge at the strainer.This information ultimately applies to both the head loss and vortex calculations,as the ECCS sump strainer and its predicted performance are analyticallydownstream of the debris quantifications. Significant conservatisms incorporatedin the design of the strainer are listed below.

McGuire ECCS Sump Strainer Head Loss Conservatisms" For the Inside the Crane Wall Enclosure structure, substantial perforated

plate area is not credited for large debris capture.

" The quantities of debris that transport to the McGuire modified ECCS sumpstrainer are conservative due to maximum transport assumptions.

Page 33 of 86



* The curbs around the refueling canal and Containment Air Return Fan pitsexisting in Upper Containment will act as large debris interceptors, but arenot credited in the transport evaluation.

" No credit is taken for debris remaining on structures and equipment abovethe pool water level.

" No credit is taken for the shielding of insulation and coatings by majorequipment in the break ZOI.

" The initial fibrous debris volume from destroyed insulation used for sizingthe modified McGuire ECCS sump strainer is based on a 17D break zone-of-influence (ZOI). As discussed in Section 3(b) of Enclosure 2, WCAP-16710-P recommends a 7D ZOI for jacketed fiber insulation based onspecific testing.

" The failed coatings debris volume used for sizing the McGuire modifiedECCS sump strainer is conservatively high.

" The assumed flowrate in the McGuire ECCS sump strainer head losscalculations is conservatively high.

" The assumed temperature in the McGuire ECCS sump strainer head losscalculations is conservatively low.

McGuire ECCS Sump Strainer Vortex Evaluation Conservatisms" As discussed in item 3(f)(3) above, a range of approach velocities were

tested in the vortex suppression evaluation; the highest strainer approachvelocities tested were higher than the nominal velocities predicted to occurin the McGuire ECCS sump by a factor of three or more. The vortexsuppressor successfully eliminated the vortices at all tested approachvelocities.

" The water level during the vortex suppression evaluation was maintainedonly 3 inches above the top surface of the top-hat perforated plate (theexpected minimum water level above the top-hat perforated plate atMcGuire is about 4 inches).

Page 34 of 86



3(f)(9) Provide a summary of the methodology, assumptions, bases forthe assumptions, and results for the clean strainer head losscalculation.

McGuire Response:

MethodologyThe following methodology-is used to calculate the Clean Strainer Head Loss(CSHL):

1. The strainer net surface area was determined. The net area is defined as thetop-hat perforated plate surface area less the area which is unable to takeflow due to blockage by stiffener rings, solid margins, or other structural steel.

2. The head loss through a single top-hat is calculated, using data from the Top-hat Array testing.

3. The head loss due to the flow traveling through the plenum was calculated.

4. In order to estimate head loss through the plenum, a hydraulic diameter wascalculated for each section with a unique cross sectional area.'

5. The largest head loss experienced by a top-hat for its respective flowcondition and the plenum head loss was summed to produce the mostconservative clean strainer head loss.

6. The Clean Strainer Head Loss is calculated for both Train A and Train Bsuction lines of the Emergency Core Cooling System, which are suppliedrecirculation water through redundant headers. The McGuire ECCS sumpstrainer consists of five major sections per train: the Extension header, theInside the Crane Wall Section (wing plenums), the 18-inch pipe header, theOutside the Crane Wall (OCW, or Pipechase) Section, and the OCW middleheader. Significant cases considered when calculating the Clean StrainerHead Loss are:

" Two-train RCS Cold Leg recirculation with Safety Injection to RCSHot Leg (Higher flows with Safety Injection aligned to Hot Leg asopposed to Cold Leg)

* ECCS Recirculation with Safety Injection to RCS Hot Leg; RHR toCharging/SI isolation valve (ND-58A) closed; "A" train RHR notoperating

Assumptions/BasesThe following assumptions are made for the Clean Strainer Head Losscalculation:

1. Steady, incompressible flow is assumed. By definition, the system is water-solid and single-phase.

Page 35 of 86

Enclosure 2Information Addressing Issues Identified in Staff Content. Guide for Generic Letter


2. The lowest sump water temperature is assumed to be constant at 600F. Fordynamic head losses, this is conservative.

3. The containment pressure is assumed to be 14.7 psia. This assumption isreasonable because the water properties associated with pressure are notsignificantly affected by the pressure term.

4. The head loss across the strainer top-hat modules (including the knitted wiremesh bypass eliminator feature) as a function of the approach velocity isdetermined by prototype array testing.

5. Head losses associated with minor obstructions in the flow path are assumedto be negligible.

6. The effective roughness for commercial steel pipe is used for the stainlesssteel plenum.

7. The modeling of the enclosure is done using a flow path with conservativelosses. Relative to the overall head loss of the strainer, the numbers utilizedare small. Thus, using an approximate flow path is reasonable.

McGuire Clean Strainer Head LossThe McGuire clean strainer head loss, based on the installed Unit 1 strainer area,,is calculated as 5.3 feet of water for the maximum recirculation flow condition,and 3.54 feet of water for single train RHR/two-train CS operation.

The McGuire Unit 2 ECCS sump strainer has not yet been completed; it will befully installed in spring 2008. The Unit 2 strainer, when completed, will beapproximately the same size as the installed McGuire Unit 1 strainer.

3(f)(10) Provide a summary of the methodology, assumptions, bases forthe assumptions, and results for the debris head loss analysis.

McGuire Response:The responses provided in items 3(f)(4), 3(f)(5), 3(f)(7), 3(f)(8) and 3(f)(9) of thissection address the methodology, assumptions, bases for assumptions, andresults for the McGuire ECCS sump strainer debris bed head loss analysis.

Page 36 of 86



3(f)(1 1) State whether the sump is partially submerged or vented (i.e.lacks a complete water seal over its entire surface) for anyaccident scenarios and describe what failure criteria in addition toloss of net positive suction heard (NPSH) margin were applied toaddress potential inability to pass the required flow through thestrainer.

McGuire Response:As discussed in the response to RAI #38 of Enclosure 1, there are no vents orother penetrations through the modified McGuire ECCS strainer connecting theinterior of the strainer to the containment atmosphere above the containmentminimum water level. The McGuire strainer is designed to be fully submerged,and as discussed in the response to RAI #41 of Enclosure 1, is fully submergedeven in the bounding SBLOCA scenario.

3(f)(12) State whether near-field settling was credited for the head-losstesting and, if so, provide a description of the scaling analysisused to justify near-field credit.

McGuire Response:Upstream debris settling due to the "near-field effect" is not credited in the headloss testing or in the analytical design basis of McGuire's modified ECCS sumpstrainers.

3(f)(1 3) State whether temperature/viscosity was used to scale the resultsof the head loss tests to actual plant conditions. If scaling wasused, provide the basis for concluding that boreholes or otherdifferential-pressure induced effects did not affect themorphology of the test debris bed.

McGuire Response:

The top-hat array testing that generated the head loss correlations for theMcGuire ECCS sump strainer top-hats was performed at room temperature,which required that a temperature coefficient be used to scale the head lossresults to plant conditions.

As discussed in item 3(f)(8) above, the debris generation and debris transportcalculations used to size the McGuire modified ECCS sump strainer producedconservative debris loads. These debris loads were incorporated into theprototype top-hat array test, which was performed to evaluate the top-hatperformance under various debris loading conditions.

The prototype array thick bed testing demonstrated the bridging of fibrous debrisbetween the top-hat strainer modules arranged in a 2 x 3 array (i.e., filling in the

Page 37 of 86



interstitial volume). The intent of the testing was to show that this bridging, andthe subsequent uniform debris loading of the modules, resulted in higher headloss than the thin bed test scenarios. Under these conditions, the interstitialvolume of the top-hats is completely filled with fibrous debris, and no evidence ofanomalous debris bed formation, including boreholes or other differential-pressure induced effects, was observed that was attributed to the testtemperature.

The Duke Integrated Prototype Test (IPT) for chemical effects, described in theresponse to RAI #11 of Enclosure 1, used the predicted McGuire post-DBAcontainment sump pool temperature cool-down profile, so no temperaturecoefficient is necessary.

3(f)(14) State whether containment accident pressure was credited inevaluating whether flashing would occur across the strainersurface, and if so, summarize the methodology used to determinethe available containment pressure.

McGuire Response:Containment DBA pressure is not credited in the modified McGuire ECCS sumpstrainer design.

Page 38 of 86

Enclosure 2'Information Addressing Issues Identified in Staff Content Guide for Generic Letter


3(g) Net Positive Suction Head (NPSH)

The objective of the NPSH section is to calculate the NPSH margin for theECCS and CSS pumps that would exist during a loss-of-coolant accident(LOCA) considering a spectrum of break sizes.

3(g)(1) Provide applicable pump flow rates, the total recirculation sumpflow rate, sump temperature(s), and minimum containment waterlevel.

McGuire Response:The McGuire ECCS/CS pump alignment in the containment sump poolrecirculation mode requires that the Residual Heat Removal System pumps,taking suction from the sump pool, supply both the Safety Injection and ChargingSystem pump inlets to ensure adequate NPSH is available. The ContainmentSpray System pumps take flow from the containment sump pool in recirculationmode as well.

Table 3G1-1 below lists the applicable flowrates for the McGuire RHR/CSpumps. The flowrates given in this table are representative flowrates for bothUnit 1 and Unit 2.

Table 3G1-1McGuire RHR / CS Pump Flow Rates

Flow at Available Flow at RequiredNPSH NPSH

Residual Heat RemovalPump Flow 4819 gpm 5000 gpmContainment Spray PumpFlow 3670 gpm 4000 gpm

Other information requested follows.

* Total ECCS Sump Pool Recirculation Flow Rate: The limiting NPSH marginfor the CS pumps and RHR pumps is two-train RHR/CS recirculationoperation with SI to the RCS Hot Leg and the RHR to Charging/SI isolationvalve closed, which results in a total analyzed recirculation flow of 15,700gpm. Additionally, the single train RHR/two-train CS operation results in thelimiting NPSH for the CS pumps which results in a recirculation flow of12,100 gpm.

" ECCS Sump Pool Temperatures: 190 0F, decreasing to 900F. As discussedin the response to RAI #11 of Enclosure 1, 190°F is the peak temperature at

Page 39 of 86



the beginning of ECCS sump pool recirculation; the pool temperaturedeclines rapidly after initiation of recirculation, ultimately reaching 90'F atthe end of the ECCS mission time.

ECCS Sump Pool Minimum Water Level: As discussed in the response toRAI #41 of Enclosure 1, the limiting analytical case for minimum ECCSsump level at McGuire is characterized as a SBLOCA during whichContainment Spray does not actuate and there is no water sourcecontribution from ice melt. For this SBLOCA scenario, a sump pool level of36 inches above the sump floor is calculated, which represents asubmergence level of at least 2 inches above the top of the vortexsuppressors.

3(g)(2) Describe the assumptions used in the calculations for the aboveparameters and the sources/bases of the assumptions.

McGuire Response:

General Assumptions* No credit is taken for increased containment pressure during an accident.

* Containment ECCS sump pool temperature is 1907F.

" ECCS sump minimum level (floor) elevation of 725'+0" is used to determinethe available NPSH.

* The SI/Charging pump required NPSH is based on single pump run-out flowrequirements.

* RHR/CS pump required NPSH is taken at a flowrate above that achievablebased on system resistance.

* The hydraulic model was based on a Unit 1 model which is assumed to alsobe applicable to Unit 2. This assumption is supported by:

1. Similar ECCS/CS pump hydraulic capability.

2. The most limiting pump NPSH required value was used for theacceptance criteria.

3. The overall system configuration/resistance and flowrates are similar,with the exception of the 2B CS Heat Exchanger (HX). The 2B HX hasspray flow on the shell side, such that the overall resistance coefficientis lower. The B train CS flow model used the lower resistancecoefficient for the 2B CS HX to conservatively model higher flow rate.

4. Suction piping configuration differences and resultant losses arejudged to be insignificant.

Page 40 of 86



ECCS Sump Pool Temperature AssumptionsSee the response to RAI #11 of Enclosure 1 for the model and assumptions usedin generating this temperature profile.

ECCS Sump Pool Minimum Water Level AssumptionsSee the response to RAI #41 of Enclosure 1 for the model and assumptions usedin generating the minimum sump pool water level.

3(g)(3) Provide the basis for the required NPSH values, e.g., threepercent head drop or other criterion.

McGuire Response:The required NPSH values for the ECCS/CS pumps are taken from theapplicable pump head curves. In determining required NPSH values, flowratesbeyond the limiting flow rates for the hydraulic model were conservatively used.

3(g)(4) Describe how friction and other flow losses are accounted for.

McGuire Response:The methodology and assumptions used for the hydraulic modeling of themodified McGuire ECCS sump strainer are located in Section 3(f), item 3(f)(9).For the remainder of the connected ECCS/CS piping systems, hydraulic modelsare generated using standard methodologies which apply appropriate resistancecoefficients and friction factors (e.g., the ECCS/CS NPSH calculations include arepresentative piping roughness factor based on commercial steel piping).

3(g)(5) Describe the system response scenarios for LBLOCA andSBLOCAs.

McGuire Response:Upon initiation of a LBLOCA or SBLOCA, the ECCS/CS systems respond asdescribed following:

Large Break LOCAThe ECCS will automatically start and align for Injection Phase upon receipt of aSafety Injection signal. The CSS will automatically start on high ContainmentPressure. During the Injection Phase, water is taken from the Refueling WaterStorage Tank and injected into the RCS through the cold legs. Dependent uponbreak size, the Cold Leg Accumulator tanks will also discharge into the RCS.

Upon reaching the Refueling Water Storage Tank (FWST) low level setpoint, theCold Leg Recirculation Phase is entered, where the RHR pumps and CSSpumps take suction from the containment ECCS sump pools. The RHR pumps

Page 41 of 86



then supply flow to the SI and Charging pump inlets. Only one train of RHR isrequired for the ECCS sump pool recirculation alignment.

To help control containment pressure after the ice beds have melted, one train ofauxiliary containment spray from the RHR System is also initiated.

At approximately 9 hours into the accident, the Hot Leg Recirculation Phase isentered where the ECCS pumps supply flow to both Hot and Cold Leg injectionlines (SI to the RCS Hot Legs and RHR to the RCS Cold Legs). SI pump flow tothe cold legs is isolated.for hot leg recirculation.

Small Break LOCAFor SBLOCAs, the break size determines ECCS/CS involvement. If the break issmall, Charging System flow will provide make-up from the FWST until the plantis stabilized. CSS pumps will likely not be needed for control of containmentpressure, and recirculation from the ECCS sump pool would also not beexpected. For larger SBLOCA scenarios, SI will also initiate, most likely on a lowRCS pressure signal, and take flow from the FWST. For breaks of this size theice condenser melt water will provide ECCS sump inventory, along with the RCSbreak flow and FWST contribution. ECCS recirculation will be initiated only whenthe appropriate ECCS sump pool level setpoint is reached. CS will be initiatedonly if necessary, taking suction from the ECCS sump pool when the appropriate(higher) ECCS sump pool level setpoint is reached.

3(g)(6) Describe the operational status for each ECCS and CSS pumpbefore and after the initiation of recirculation.

McGuire Response:During a LBLOCA, prior to the initiation of ECCS sump pool recirculation andduring the Cold Leg Injection mode, the ECCS and CSS pumps will be operatingas follows:

" Both RHR pumps will be running, aligned to the FWST.

" Both CS pumps will be running, aligned to the FWST.

" Both SI pumps will be running, aligned to the FWST.

" Both Charging pumps will be running, aligned to the FWST.

ECCS sump recirculation mode is initiated from decreasing FWST levelindication. As the initiation of ECCS sump pool recirculation approaches, theECCS and CSS pumps are realigned and operate as follows:

Page 42 of 86



The ECCS sump valves automatically open at FWST Low Level Setpointaligning to RHR suction. The RHR suction valves to the FWST aremanually closed. RHR pumps remain running.

* Charging and SI pumps are manually aligned to the RHR pump dischargewith both pumps running. Charging/SI pump suction from the FWST ismanually isolated.

At the FWST low-low level alarm, the CS pumps are manually secured. TheCS pump suction is realigned to the ECCS sump and both CS pumps arerestarted.

3(g)(7) Describe the single failure assumptions relevant to pumpoperation and sump performance.

McGuire Response:The McGuire ECCS components are designed such that the Charging pumps, SIpumps, RHR pumps, and CS pumps, together with their associated valves andpiping, will assure adequate core cooling in the event of a Design Basis Loss ofCoolant Accident.

For ECCS systems described below, the "short term" ends when the ECCS isplaced in ECCS recirculation mode.

Charging pumpsThe ECCS portion of the McGuire Charging System has two redundant trains ofactive and passive safety-related equipment that meet single failure criteria. Thisequipment is designed to tolerate an active failure during the short term or apassive failure in the long term following a Design Basis Accident.

Safety Injection pumpsWith two redundant trains per Unit, the McGuire SI system is designed to toleratea single active failure during the short term or an active or passive failure duringthe long term following a Design Basis Accident.

Residual Heat Removal pumpsThe McGuire RHR system is designed to meet single failure criteria, with tworedundant trains per unit. The RHR system is designed to tolerate a single activefailure during the short-term or a passive failure during the long term following aDesign Basis Accident.

Containment Spray pumpsThe McGuire Containment Spray System, including required auxiliary systems, isdesigned such that it will tolerate a single active failure during the injection phaseor a single active or passive failure during the recirculation phase following a

Page 43 of 86



Reactor Coolant System failure, without loss of its protective function. Systemactive components are redundant. System piping located within the Containmentis redundant and separated in arrangement unless fully protected fromconsequential damage which may follow any Reactor Coolant System pipefailure. Emergency power system arrangements assure the proper functioning ofthe Containment Spray System during loss-of-power conditions.

ECCS Sump StrainerAt McGuire, a single, shared (non-redundant) strainer is utilized. The need tomaintain two physically separated ECCS containment sumps or ECCS/CSS trainseparation within the same sump is unnecessary.

This is described in Duke letter to USNRC dated March 8, 2007 "LicenseAmendment Request Revising McGuire Units 1 and 2 Updated Final SafetyAnalysis Report Commitments to USNRC Regulatory Guide 1.82, Revision 0,"Sumps For Emergency Core Cooling and Containment Spray Systems" andRevising McGuire Units 1 and 2 Technical Specification SurveillanceRequirement (SR) 3.5.2.8 and Associated Bases.

3(g)(8) Describe how the containment sump water level is determined.

McGuire Response:Two containment water level indicator channels provide the Control Room withECCS sump water level indication. Additionally, two level switches are providedto annunciate when realignment to the ECCS sump is allowable for the ECCSand CS pumps.

3(g)(9) Provide assumptions that are included in the analysis to ensure aminimum (conservative) water level is used in determining NPSHmargin.

McGuire Response:As noted in item 3(g)(2) above, available NPSH is determined from the floorelevation of the McGuire ECCS sump (elevation 725' + 0") instead of theavailable water level. Using the floor elevation of the ECCS sump is aconservative assumption in calculating NPSH margin.

Page 44 of 86



3(g)(10) Describe whether and how the following volumes have beenaccounted for in pool level calculations: empty spray pipe, waterdroplets, condensation and holdup on horizontal and verticalsurfaces. If any are not accounted for, explain why.

McGuire Response:A SBLOCA is the limiting case for pool minimum volume calculations. Indetermining applicable inventory penalties for SBLOCAs, the following diversionswere accounted for:

* Incore Room Diversion" Volume Control Tank (VCT) Diversion* Pressurizer Relief Tank (PRT) Diversion• Lower Containment Ventilation (VL) Diversion

The following volumes were accounted for when determining Upper ContainmentHoldup Volume:

* Refueling Canal Holdup* Refueling Deck Holdup (3-inch curb around the refueling canal)• CS System Piping Volume* Airborne spray volume (droplets) in Upper Containment* Water held up in containment draining down walls

3(g)(11) Provide assumptions (and their bases) as to what equipment willdisplace water resulting in higher pool level.

McGuire Response:Various types of large robust miscellaneous equipment (e.g., tanks, housings,piping, base plates, rupture restraints, and supports) were assumed to displacewater in the determination of ECCS sump volume. Smaller miscellaneousequipment (e.g., small bore piping, cable trays, transformers, and tubing) wereconservatively excluded. The ECCS sump strainer is also included as a volumedisplacer. In addition, the calculation conservatively does not take waterdisplacement credit for insulation around pipes.

3(g)(12) Provide assumptions (and their bases) as to what water sourcesprovide pool volume and how much volume is from each source.

McGuire Response:The following conservative initial condition assumptions are made in determiningthe volume of water available for the ECCS sump inventory for any break sizeinside Containment:

Page 45 of 86



" Refueling Water Storage Tank (FWST): The credited volume for the FWSTis the Technical Specification minimum allowable volume. EmergencyProcedure guidance directs that ECCS pumps be realigned to recirculationmode at the FWST low-low level indication. This level is error-adjustedupward to conservatively reflect maximum remaining tank volume andminimize injected volume.

" Reactor Coolant System: The initial Reactor Coolant Hot Full Power Massis converted to the Reactor Coolant Mass at 200°F and 300 psia for breakconditions.

" Ice Condenser Ice Bed Inventory: The ice bed total mass is assumed to beat the technical specification minimum.

" Cold Leg Accumulators: For larger break sizes, the Technical Specificationnominal inventory volume in the Cold Leg Accumulators is assumed todischarge into the RCS.

3(g)(13) If credit is taken for containment accident pressure in determiningavailable NPSH, provide description of the calculation ofcontainment accident pressure used in determining the availableNPSH.

McGuire Response:No credit is taken for containment accident pressure in the McGuire ECCS/CSpump NPSH calculations.

3(g)(14) Provide assumptions made which minimize the containmentaccident pressure and maximize the sump water temperature.

McGuire Response:The following significant analytical conditions related to containment accidentpressure and temperature response are assumed at-the onset of a postulatedLBLOCA event:

* Containment pressure is assumed to be atmospheric. No credit is taken forcontainment accident pressure in the McGuire ECCS/CS pump NPSHcalculations.

* FWST water inventory temperature is assumed to be at the technicalspecification maximum.

* FWST inventory is assumed to be at its technical specification minimum.

• Nuclear Service Water (ultimate heat sink) temperature is assumed to beconservatively high.

" Ice bed inventory is assumed to be at the technical specification minimum.

Page 46 of 86



* Minimum safeguards are assumed

3(g)(15) Specify whether the containment accident pressure is set at thevapor pressure corresponding to the sump liquid temperature.

McGuire Response:No credit is taken in ECCS/CS pump NPSH calculations for increasedcontainment pressure during an accident. The vapor pressure input to theavailable NPSH hydraulic model for the ECCS/CS pumps is taken at the sumppool maximum temperature of 190°F at atmospheric pressure.

3(g)(16) Provide the NPSH margin results for pumps taking suction fromthe sump in recirculation mode.

McGuire Response:The limiting NPSH margin for the RHR/CS pumps taking suction from the ECCSsump pool in recirculation mode is shown in Table 3G16-1.

The limiting NPSH available values shown in the Table assume two-trainoperation to maximize the suction flow path head losses; and no ECCS sumpstrainer differential pressure losses. Strainer differential pressure lossesgenerally are time dependent and largely offset by vapor pressure reduction andsump pool level increase over time. The refined debris-loaded head losspredicted at the ECCS sump strainer, including chemical effects, is discussed inSection 3(f) of Enclosure 2.

Table 3G16-1

McGuire RHR / CS Pump NPSH Margins

NPSH NPSH NPSHRequired Available Margin(ft-water) (ft-water) (ft-water)

CS Pumps 19 1 > 29 1> 10

RHRPumps 19 > 33 >14

As described in the response to RAI #12 of Enclosure 1, the total ECCS sumpstrainer head loss for the McGuire strainers will be based on the results of theIntegrated Prototype Test (IPT). The IPT is more fully described in the responseto RAI #11 of Enclosure 1. Upon completion of the IPT documentation, a refineddebris load head loss calculation will incorporate any added consequence oftested chemical effects. This is a commitment identified in Duke letter dated

Page 47 of 86



November 6, 2007 "Request for Extension of Completion Dates for McGuireUnits 1 and 2 Corrective Actions Required by NRC Generic Letter (GL) 2004-02".

Final limiting NPSH margins for the McGuire RHR/CS pumps will be determinedbased on the refined debris-loaded head loss calculation results and reported inthe response to RAI #12 of Enclosure 1. Duke expects to have thisdocumentation and to supplement the response to RAI #12 by April 30, 2008.

Page 48 of 86



3(h) Coatings Evaluation

The objective of the coatings evaluation section is to determine the plant-specific ZOI and debris characteristics for coatings for use in determiningthe eventual contribution of coatings to overall head loss at the sumpscreen.

3(h)(1) Provide a summary of type(s) of coating systems used incontainment, e.g., Carboline CZ 11 Inorganic Zinc primer, Ameron90 epoxy finish coat.

McGuire Response:The qualified coatings systems for the concrete surfaces and the steel structuresand components inside containment consist of:

* Carboline 890* Valspar 76-Series High Build* Valspar 89-C-3-00* Valspar 13-F-12KR-00 MZ #7 Primer* Valspar 89-Series Epoxy* Carbozinc 11 SG Primer

The unqualified coatings inside containment consist of:

* Epoxy* Alkyd Enamel* Cold galvanizing

3(h)(2) Describe and provide bases for assumptions made in post-LOCApaint debris transport analysis.

McGuire Response:The following assumptions relating to failed coatings debris are made for thedebris transport analysis:

* It is assumed that the settling velocity of fine debris (including paintparticulate) can be calculated using Stokes' Law. This is a reasonableassumption since the particulate debris is generally spherical and wouldsettle slowly (within the applicability of Stokes' Law).

* The unqualified coatings in containment are assumed to be uniformlydistributed in the containment pool at the beginning of recirculation. This isa reasonable assumption since the unqualified coatings are scattered in

Page 49 of 86



small quantities throughout containment. The assumption for distribution isnot significant since particulate is so readily transported.

* The qualified coatings are assumed to be Carboline 890 since this systemhas the largest qualified coating thickness.

" In accordance with the NRC SER of NEI 04-07 guidance, all unqualifiedcoatings in containment are assumed to fail, as well as all qualified coatingswithin the break ZOI.

* It is assumed that failed coatings in upper containment are washed down bycontainment sprays. 100% of paint fines located in the ice condenser andupper containment are assumed to washdown and transport to the strainer.

* All failed coatings (including coatings inside the break ZOI and unqualifiedcoatings outside the break ZOI), are conservatively assumed to beparticulate. No coating debris in the size/shape of paint fines or paint chipsis considered.

3(h)(3) Discuss suction strainer head loss testing performed as it relatesto both qualified and unqualified coatings and what surrogatematerial was used to simulate coatings debris.

McGuire Response:Two sets of debris head loss tests were performed for the McGuire ECCS sumpstrainer design. A Top-hat Array Test, utilizing horizontally positioned top-hatstrainer modules in a 2 x 3 arrangement, was conducted to determine thesusceptibility of the top-hats to the thin-bed effect and to determine the head losscorrelation using representative debris loading challenges, including particulates.This testing is also described in the response to RAI #40 of Enclosure 1. Asurrogate, SIL-CO-SILTM 53 Ground Silica was used for the failed qualified andunqualified coatings in the Top-hat Array Test.

Subsequently, an Integrated Prototype Test was performed utilizing onehorizontally positioned top-hat module, to determine actual head loss during theECCS mission time using refined debris loading, including particulates, andchemical effects. This testing is described further in the resyonse to RAI #11 ofEnclosure 1. For this test, silica oxide flour (1250 Novacite ) represented thefailed qualified and unqualified coatings. No surrogate for failed inorganic zinccoatings was used in this test, since McGuire contains the bounding particulateload for the Duke ice condenser plants and has no inorganic zinc coatings.

In the containment sump pool, the particulate material will occupy a certainvolume in the fibrous debris space resulting in increased resistance to flow andhigher head loss. The surrogate material volume was therefore adjusted in both

Page 50 of 86



of these tests to match the volume of the failed coatings particulate for coatingsthat are less dense than the surrogate.

3(h)(4) Provide bases for the choice of surrogates.

McGuire Response:SIL-CO-SIL 53 Ground Silica is the surrogate used to represent the failedqualified and unqualified coatings at McGuire in the Top-hat Array Test. Theground silica material specific gravity is 2.65, which corresponds to a density of165 lb/ft3 ; epoxy and alkyd coatings densities range from 94 Ib/ft3 to 98 lb/ft3 perNEI 04-07 guidance. The ground silica is a spherical particulate ranging in sizefrom just under 1 micron to approximately 100 microns. The majority of the failedcoatings are on the order of 10 microns in size or greater. Since a significantportion of the ground silica material is less than 10 microns, the ground silicawould tend to produce a debris bed with a lower porosity and higher surface-to-volume ratio than a debris bed comprised of failed coating material alone. Thus,the use of ground silica as a surrogate for failed coating debris in the Top-hatArray Test is conservative.

Silica oxide flour (1250 Novacite®) represented the failed qualified andunqualified coatings in the Integrated Prototype Test. This particulate debris.surrogate material. was selected based on chemical reactivity and a comparisonof the microscopic densities of the material. Epoxy and alkyd coatings densitiesrange from 94 Ib/ft3 to 98 lb/ft3 per NEI 04-07 guidance 15; silica oxide flour has amaterial specific gravity of 2.65, corresponding to a microscopic density of 165Ib/ft3. The particle size for failed epoxy and alkyd coatings is assumed to be 10microns. The silica oxide flour surrogate material is a spherical particulate where99% is less than 45 microns in diameter and 69% is less than 10 microns.

3(h)(5) Describe and provide bases for coatings debris generationassumptions. For example, describe how the quantity of paintdebris was determined based on ZOI size for qualified andunqualified coatings.

McGuire Response:When considering the quantification of post-accident coatings debris generation,the guidelines presented in the NRC SER of NEI 04-07 were followed. Per NEI04-07, qualified and unqualified coatings within the ZOI are assumed to fail as aresult of impingement and post-accident environmental conditions. Qualifiedcoatings outside the ZOI are assumed to remain intact and adhered to theirsubstrate.

A CAD model of containment is used to determine the area of qualified coatingswithin the ZOI for each break in consideration. The volume of qualified coatings

Page 51 of 86



within the ZOI is calculated based on a maximum of 12 mils thick for concretefloors and walls and a maximum of 11 mils thick for steel surfaces.

The following assumptions are made in the McGuire debris generation evaluationrelated to coatings:

" Per NEI 04-07 guidance, all unqualified coatings in containment areassumed to fail as 10 micron spheres during the HELB; the containmentwalkdown reports are used as the basis for the quantity of unqualifiedcoatings.

* Qualified coatings in the break ZOI are also assumed to fail as 10 micronspheres. The walkdown report and coatings specifications were used todetermine the type, thickness, and number of coats applied.

* Qualified coatings are assumed to fail within a 5D ZOI as defined by theWCAP-16568-P methodology, in lieu of the 10D ZOi defined by NRC SERof NEI 04-07.

* Unqualified epoxy coatings debris quantities and transport metrics arerefined based on analysis of OEM coatings performed by EPRI. It isassumed that Duke-applied coatings inside containment are similar to themanufacturer-applied coatings used in the analysis. Unqualified alkydcoatings debris and qualified coatings debris are not affected by thisrefinement.

* The qualified coatings are assumed to be Carboline 890 since this systemhas the largest qualified coating thickness.

Other assumptions related to the transport of failed coatings debris are identifiedin the response to item 3(h)(2) above.

Table 3H5-1 below shows the ZOI radius and destruction pressure for McGuirequalified coatings.

Page 52 of 86



Table 3H5-1

ZOI Radius for McGuire Qualified CoatingsDebris Type Destruction Pressure ZOI Radius/Break

(psi) Diameter(LID)

Protective Coatings(epoxy and epoxy-phenolic Not measured** 5.0paints)

** The approach taken for testing was to position the test coupon a distance from the jet andobserve the coatings performance. If no degradation of coatings was observed, a ZOI wascalculated using the ANSI/ANS 58.2-1988 jet expansion model. A specific destruction pressurewas not measured.

Postulated qualified and unqualified coatings debris quantities are located in theresponse to item 3(h)(6) below.

3(h)(6) -Describe what debris characteristics were assumed, i.e., chips,particulate, size distribution and provide bases for theassumptions.

McGuire Response:All failed coatings (including coatings inside the break ZOI and unqualifiedcoatings outside the break ZOI), are conservatively assumed to be particulate.No coating debris in the size/shape of paint fines or paint chips is considered.

The NEI 04-07 guidance report and companion SER are followed for the coatingsdebris evaluation, and conservatively assume that the coatings will all fail ashighly transportable 10 micron spherical particles. The qualified coatingmaterials at McGuire are a maximum of 12 mils thick for concrete floors andwalls and a maximum of 11 mils thick for steel surfaces. It is conservative toassume all of this coating material will erode to pigment-sized particles. Further,qualified coatings are assumed to be Carboline 890 since this system has thelargest qualified coating thickness.

The debris characteristics and postulated debris quantities for McGuire qualifiedand unqualified coatings are shown in Tables 3H6-1 and 3H6-2 below:

Page 53 of 86



Table 3H6-1

McGuire Qualified Coatin s Characteristics

Analysis Volume Density WeightArea Within ZOI Area (ft2) Thickness Size (ft3) (lb/ft3) (Ib)Concrete Surfaces 568 12 mils 10 micron 0.57 118 67.3Steel Surfaces 928 11 mils 10 micron 0.85 118 100.3Total 1,496 N/A 10 micron 1.42 118 167.6

Table 3H6-2

McGuire Unqualified Coatings Characteristics

Coating Total DFT* Volume Density Weight** AnalysisMaterial Area (ftz) (mils) (ft3) (Ib/ft3) (Ib) SizeEpoxy 13,917 6 3.8 94 357.1 10 micronAlkyd Enamel 1,213 1.5 0.16 98 15.7 10 micronTotal 15,130 N/A 3.96 N/A 372.8 10 micron

* DFT: Dry Film Thickness

** As identified in the response to question 3(h)(5) above, unqualified epoxy coatings debrisquantities and transport metrics are refined based on analysis of OEM coatings performed byEPRI. Duke has revised the initial statistical assessment and incorporated a more conservativeunqualified particulate refinement based upon applying a 2 standard deviation correction from themean. It is assumed that Duke-applied coatings inside containment are similar to themanufacturer-applied coatings used in the analysis. Unqualified alkyd coatings debris andqualified coatings debris are not affected by this refinement.

3(h)(7) Describe any ongoing containment coating condition assessmentprogram.

McGuire Response:The comprehensive Duke Energy Corporation Containment CoatingsAssessment Program in effect at McGuire is used to identify degradedqualified/acceptable coatings and determine the amount of debris that will resultfrom these coatings. This program also ensures that qualified/acceptablecoatings remain in compliance with plant licensing requirements for design-basisaccident (DBA) performance.

This assessment program is discussed in detail in the response to RAI #25 ofEnclosure 1.

Page 54 of 86



3(i) Debris Source Term Refinements

The objective of the debris source term section is to identify any significantdesign and operational measures taken to control or reduce the plantdebris source term to prevent potential adverse effects on the ECCS andCSS recirculation functions.

Provide the information requested in GL 04-02 Requested Information Item2(f) regarding programmatic controls taken to limit debris sources incontainment.

GL 2004-02 Requested Information Item 2(f)A description of the existing or planned programmatic controls that will ensurethat potential sources of debris introduced into containment (e.g., insulations,signs, coatings, and foreign materials) will be assessed for potential adverseeffects on the ECCS and CSS recirculation functions. Addressees may referencetheir responses to GL 98-04, "Potential for Degradation of the Emergency CoreCooling System and the Containment Spray System after a Loss-of-CoolantAccident Because of Construction and Protective Coating Deficiencies andForeign Material in Containment," to the extent that their responses addressthese specific foreign material control issues.

In responding to GL 2004-02 Requested Information item 2(f), provide thefollowing:

3(i)(1) A summary of the containment housekeeping programmaticcontrols in place to control or reduce the latent debris burden..Specifically for RMI/low-fiber plants, provide a description ofprogrammatic controls to maintain the latent debris fiber sourceterm into the future to ensure assumptions and conclusionsregarding inability to form a thin bed of fibrous debris remainvalid.

McGuire Response:McGuire is not considered a low fiber plant, so the latent debris (particulate)source term is not dominant. The control of latent debris and miscellaneouslatent debris (tags, labels, etc.) is still important and essential, and McGuire hasimplemented programmatic controls to ensure that potential sources of debristhat may be introduced into containment will be assessed for adverse effects onthe ECCS and Containment Spray recirculation functions.

Page 55 of 86

Enclosure 2Information Addressing Issues Identified in Staff Content Guide. for Generic Letter

2004-02 Supplemental ResponsesMcGu ire Nuclear Station

These programmatic controls include:

* Coatingis Progiram:Duke has established controls for procurement, application, andmaintenance of Duke-applied, Service Level 1 protective Coatings usedinside containment. Further discussion regarding Duke's ContainmentCoatings Assessment Program is located in the response to RAI #25 ofEnclosure 1.

* Containment Housekeepingi/Materiel Condition:Extensive containment cleaning is performed during each refueling outageusing water spray, vacuuming and hand wiping. Localized wash downs areperformed as needed and visual inspections are performed on the remainingareas of containment. Foreign material is removed as necessary.Upgrades to existing foreign material control procedures require materialaccountability logs to be maintained in Modes 1 through 4 for items carriedinto and out of Containment. These controls are implemented usingadministrative procedures.

" The plant labeling process has been enhanced to require that any additionallabels or signs placed inside containment be evaluated to ensure that thedesign basis for transportable debris is not invalidated.

" McGuire Technical Specification Surveillance Reqiuirement (SR):McGuire Technical Specification Surveillance Requirement 3.5.2.8 requiresthat the ECCS sump be visually inspected to verify there are no restrictionsas a result of debris, and no evidence of structural distress or abnormalcorrosion present prior to declaring the ECCS sump operable. A visualinspection of containment is performed to ensure no loose material ispresent which could be transported to the Containment Sump and causerestriction of the ECCS pump suction during accident conditions prior to thetransition from Mode 5 to Mode 4 operations. When these inspections areperformed, major outage work is complete, and any remaining loosematerial in containment must be logged and tracked in accordance withstation procedures for control and accountability. If any debris, damage ordeficiency were to be discovered during the inspection, station processesrequire entry into the corrective action program, with the requisiteinvestigation and implementation of appropriate corrective action prior to thetransition from Mode 5 to Mode 4.

* Additionally, McGuire Selected Licensee Commitment 16.6.1 ensures that avisual inspection is performed to identify any loose debris insidecontainment and ensure it is removed prior to establishing containment

Page 56 of 86



integrity and following entries made after containment integrity isestablished.

Modification Process:Duke's modification process currently includes an administrative procedurethat directs the design and implementation of engineering changes in theplant. This procedure directs that engineering changes be evaluated forsystem interactions. As part of this evaluation, there is direction to includeconsideration of any potential adverse effect with regard to debris sourcesand/or debris transport paths associated with the containment sump.

3(i)(2) A summary of the foreign material exclusion programmaticcontrols in place to control the introduction of foreign materialinto the containment.

McGuire Response:As noted above, visual inspections are performed in containment and foreignmaterial is removed as necessary. Upgraded foreign material control proceduresrequire material accountability logs to be maintained in Modes 1 through 4 foritems carried into and out of containment. These controls are implemented usingadministrative procedures.

The plant labeling process has been enhanced to require that any additionallabels or signs placed inside containment be evaluated to ensure that the designbasis for transportable debris is not invalidated.

I-

3(i)(3) A description of how permanent plant changes insidecontainment are programmatically controlled so as to not changethe analytical assumptions and numerical inputs of the licenseeanalyses supporting the conclusion that the reactor plant remainsin compliance with 10 CFR 50.46 and related regulatoryrequirements.

McGuire Response:Duke's modification process currently includes an administrative procedure thatdirects the design and implementation of engineering changes in the plant. Thisprocedure directs that engineering changes be evaluated for system interactions.As part of this evaluation, there is direction to include consideration of anypotential adverse effect with regard to debris sources and/or debris transportpaths associated with the containment sump.

Page 57 of 86



3(i)(4) A description of how maintenance activities including associatedtemporary changes are assessed and managed in accordancewith the Maintenance Rule, 10 CFR 50.65.

McGuire Response:Risk management per 1OCFR50.65 a(4) at McGuire is managedprogrammatically during both innage periods and outage periods, as describedfollowing:

Operational Risk Management (Modes 1-3) per IOCFR 50.65 a(4)

To ensure compliance with 1OCFR 50.65 a(4), risk assessments are performedprior to conducting any maintenance at McGuire. Maintenance includes allactivities traditionally associated with identifying and correcting degradedconditions including corrective maintenance, plant Engineering Changes, andpreventive maintenance including surveillance, predictive and preventiveactivities.

Temporary alterations are maintenance-related activities that do not permanentlyalter the design or design function of plant structures, systems, or components(SSCs). NEI 96-07, "Guidelines for 10CFR 50.59 Implementation", includesdiscussion to advise between three distinct but related topics: MaintenanceRule, Maintenance Activities, and Temporary Alterations. Compliance with10CFR 50.65 a(4), Maintenance Rule, requires any temporary alteration to beevaluated for risk prior to performing the work. Once these alterations are inplace, they may exist for ninety days of power operation before they must beconsidered as potentially being a permanent Engineering Change.

Since the temporary alterations are associated with maintenance activities, noreview is required under 1OCFR 50.59 unless the measures are expected toremain in place for greater than ninety days of power operation. If, during poweroperation, the temporary alteration is expected to be in effect for greater thanninety days, the temporary alteration is screened and if necessary evaluationperformed under 1OCFR 50.59 prior to implementation.

Shutdown Risk Management (Modes 4, 5, 6, and No-Mode) per IOCFR 50.65a(4)

Consistent with 1OCFR 50.65 a(4) requirements for outage periods, McGuiremaintenance activities during outages are cognizant of the risk associated withwork evolutions, and the out-of-service duration of risk significant componentsare managed to mitigate risk.

For activities that create temporary alterations such as lifting leads, placingjumpers on terminals, and installing trips and bypasses, the associated

Page 58 of 86



equipment is considered to be out of service. Conservatively, the SSC isconsidered to be unavailable to perform its function and is evaluated as suchduring the risk assessment.

If an SSC is required to be available with a temporary alteration in place, anevaluation of the effects of the alteration must be performed. Only afterevaluation can the SSC be determined to be available with temporary alterationsin place.

For activities that install other temporary alterations such as scaffold, leadshielding, and supports, programs are in place to evaluate and control the effectsof those alterations.

3(i)(5) If any of the following suggested design and operationalrefinements given in the guidance report (guidance report,Section 5) and SE (SE, Section 5.1) were used, summarize theapplication of the refinements.

" Recent or planned insulation change-outs in the containment whichwill reduce the debris burden at the sump strainers.

" Any actions taken to modify existing insulation (e.g., jacketing orbanding) to reduce the debris burden at the sump strainers.

" Modifications to equipment or systems conducted to reduce the debrisburden at the sump strainers.

" Actions taken to modify or improve the containment coatings program.

McGuire Response:Change-out of InsulationWhile McGuire maintains the strategic option to replace insulation, the changeout of insulation in the McGuire containments to reduce the debris burden at theECCS sump strainers is not necessary to be in full compliance with therequirements of GL 2004-02.

Modify Existing InsulationAs discussed in the response to RAI #35 of Enclosure 1, Microtherm® insulationpreviously installed on portions of the McGuire Unit 1 and Unit 2 reactor vesselheads was removed and replaced with Reflective Metal Insulation (RMI). Thereactor vessel head is not in the area of the limiting break. Debris transportcalculations show that RMI (for the limiting break location) will not transport to thecontainment sump.

Page 59 of 86



Modify Other Equipment or SystemsElectromark® labels have been evaluated as capable of withstanding the limitingbreak in all areas of containment except inside the crane wall in lowercontainment, and the miscellaneous latent debris quantification has beenadjusted accordingly.

Duke's modification process currently includes an administrative procedure thatdirects the design and implementation of engineering changes in the plant. Thisprocedure directs that engineering changes be evaluated for system interactions.As part of this evaluation, there is direction to include consideration of anypotential adverse effect with regard to debris sources and/or debris transportpaths associated with the containment sump.

Modify or Improve Coatings ProgramAs discussed in detail in the response to RAI #25 of Enclosure 1, a containmentcoatings condition assessment is conducted during each refueling outage or anyother extended outage. The containment coating condition assessment protocolconsists of a visual inspection of all readily accessible coated areas by qualifiedpersonnel. When degraded coatings are visually identified, the affected areasare documented in accordance with plant procedures. Additional nondestructiveand/or destructive examinations are conducted as appropriate to define theextent of the degraded coatings and to enable disposition of the coatingdeficiency. The guidance contained in EPRI Report 109937 is used asappropriate to disposition areas of degraded coatings when discovered,including:

1. Performance of additional in situ and/or laboratory testing of degradedcoatings,

2. Removal and replacement of degraded coatings,3. Repairing degraded coatings,4. Mitigation of accident consequences related to failure of degraded coatings,5. Leaving coating in place based on evaluation of effects of failure

(detachment) of the degraded coating on ECCS system performance, and/or6. Upgrading of indeterminate coatings.

If, after identification, degraded qualified/acceptable coatings will be left in placeduring plant operation, the degraded qualified/acceptable coatings are assumedto fail and to be available for transport to the ECCS sump. After eachcontainment coatings condition assessment, the quantity listing of degradedcoatings is updated, and the revised quantity of degraded coatings is verified tomeet the acceptance limit in the ECCS debris source term analysis.

Page 60 of 86



3(j) Screen Modification PackageThe objective of the screen modification package section is to provide a basicdescription of the sump screen modification.

3(j)(1) Provide a description of the major features of the sump screendesign modification.

McGuire Response:The McGuire ECCS sump strainer modification removes the original ECCScontainment sump screen and replaces it with a larger strainer assemblydeveloped using NEI 04-07 and other industry guidance. The major features ofthe strainer design modification include the following:

* The modified strainer assembly is constructed of robust materials and islocated both inside and outside the crane wall (see Figure 3J-1).

* The modified strainer designs were expanded to approximately 1700 squarefeet, which is the largest practical installation in the existing Unit 1 and Unit 2lower containment area (accounting for strainer submergence requirements).

* The portion of the strainer assembly located Inside the Crane Wall is housedwithin a stainless steel enclosure constructed using a structural frameworkwith platform grating covering the tops and sides. Enclosure side plating is 14gage, perforated with 3/32 inch nominal diameter holes, and the enclosuretop is solid 14 gage plate.

* The portion of the strainer assembly located outside the crane wall(Pipechase) is covered by a structural framework with platform gratingcovering the top. No solid plate top or perforated plate enclosure is included.

* Outside the crane wall, the two train-specific ECCS/CSS recirculation linesconnect directly to the main waterboxes via 18-inch diameter piping. The twopipechase waterboxes are interconnected to one another via plenums andconnected to the Inside the Crane Wall strainer assemblies by 18- inchdiameter pipes that pass through crane wall penetrations.

* The strainer assemblies consist of a series of stainless steel tubular modules(top-hats) connected by a plenum to water boxes. The top-hats areconstructed from two concentric, rolled perforated plates. The openings in theperforated plate are 3/32 inch diameter nominal. Sandwiched between theconcentric tubes of each top-hat module is a bypass eliminator, fabricatedfrom fine knitted wire. This component is designed to further filter fineentrained debris that has already penetrated the perforated top-hat exterior.

Page 61 of 86



* Horizontal vortex suppressors are installed above the top-hat assemblieslocated in the pipechase. Vortex suppression for ICW strainer assemblies isprovided by the solid top deck of the enclosure.

" The McGuire modified ECCS sump strainer assembly and enclosure arenuclear safety-related and designed to withstand safe shutdown earthquakeloadings and protected from tornado missiles by virtue of being located withinthe Containment Building which, in turn, is protected by the seismicallydesigned Reactor Building. The structures are passive assemblies (i.e., nomoving parts) qualified for the design environmental conditions of the sump.These structures are designed for the containment sub-compartmentdifferential pressures from the limiting case pressurizer surge line pipe break.

Inside theCrane Wall

ICW Enclosure and Pipechase Vortex Suppression Rack not shown for clarity

Figure 3J-1

McGuire Modified ECCS Sump Strainer

Page 62 of 86



3(j)(2) Provide a list of any modifications, such as reroute of piping andother components, relocation of supports, addition of whiprestraints and missile shields, etc., necessitated by the sumpstrainer modifications.

McGuire Response:A summary of noteworthy modifications necessitated by the McGuire Unit 1 andUnit 2 modified ECCS sump strainer installations appears below.

1. Associated with installation of modified ECCS sump strainer on Unit 1

Added a whip restraint / jet barrier for the RHR piping connected to theRCS to protect strainer components.

" The "B" Train Containment Purge Ventilation Unit was removed,. Thisremoval was required to accommodate the extension of the strainer.

* Rerouting of 1-inch Liquid Waste piping.

" Rerouting of 6-inch Component Cooling piping.

* Perforated cover plate located at the bottom drain of the refuelingcanal removed.

2. Associated with installation of modified ECCS sump strainer on Unit 2 (twoPhases). Phase 1 installation is complete; Phase 2 will be installed in spring2008.

* Modification Phase 2 will add a whip restraint / jet barrier for the RHRpiping connected to the RCS to protect strainer components

* The "B" Train Containment Purge Ventilation Unit was removed. Thisremoval was required to accommodate the extension of the strainer

Perforated cover plate located at the bottom drain of the refuelingcanal removed

Page 63 of 86



3(k) Sump Structural Analysis

The objective of the sump structural analysis section is to verify the structuraladequacy of the sump strainer including seismic loads and loads due todifferential pressure, missiles, and jet forces.

Provide the information requested in GL 2004-02 Requested Information Item2(d)(vii).

GL 2004-02 Requested Information Item 2(d)(vii)Verification that the strength of the trash racks is adequate to protect the debrisscreens from missiles and other large debris. The submittal should also provideverification that the trash racks and sump screens are capable of withstandingthe loads imposed by expanding jets, missiles, the accumulation of debris, andpressure differentials caused by post-LOCA blockage under flow conditions.

3(k)(1) Summarize the design inputs, design codes, loads, and loadcombinations utilized for the sump strainer structural analysis.

McGuire Response:Table 3K1-1 below shows the design inputs for the McGuire modified ECCSsump strainer structural calculations, including the top-hats, the strainer sectionsInside the Crane Wall and in the Pipechase, the Inside the Crane Wall StrainerEnclosure, and the Pipechase Vortex Suppression Rack.

Table 3K1-1

Design Inputs/Loads for McGuire ECCS Sump Strainer

modle* /~ Pipechase PipediDesign Input Top-hat m Sectiops ICW Enclosure Vort

SUpp.,

Temperature 300 °F 300 *F 250 *F 250 oF

4 psid solid plate,Differential Pressure 10 psid 7 psid 2.69 psid NA

perforated

Dead Weight 0.29 lb/in3 0.29 lb/in3 0.29 Ib/in 3 0.29 lb/in3

Live Load -_100 psf 100 psf

Misc. Load (Cable Tray/Conduit) -- 75 lb

,IZPA Frequency 20 Hz 20 Hz 20 Hz 20 Hz*~ Damping 2% 2% 2% 2%

, I Max SSE Horizontal Acc. 0.53 g 0.53 g 0.53 g 0.53 g

.Max SSE Vertical Acc. 0.35 g 0.35 g 0.35 g 0.35 g

* Bounding top-hat length is,45 inches for structural analysis

Page 64 of 86



Table 3K1 -2 below shows the design load combinations for the Inside the CraneWall and Pipechase Sections, the Inside the Crane Wall Enclosure, and thePipechase Vortex Suppression Rack. For this table, "Fs" represents theallowable stress in steel as specified in AISC Part 1.

Table 3K1-2

Load Combinations for McGuire ECCS Sump StrainerLoadJ Cmbibamif!ons (ICfW& riiechase Sections, ICW Enclosure, Pipechase Vortex

Load Case 1 DL (Dead Load) + CL (Construction Load) = Fs

Load Case 2 DL + OL (Normal Operating Load) + Pa (Accident Differential Pressure) = Fs

Load Case 3 Not used

Load Case 4 DL + OL + OBE = Fs

Load Case 5 DL + OL + SSE = 1.5 Fs

Load Case 6 DL + OL + SSE + Ta (Accident Thermal Load) = 1.5 Fs

Load Case 7 DL + OL + SSE + Pa +Y (Pipe Rupture) = 1.5 Fs

The load combination used for the Top-hat structural calculations is Dead Weight+ SSE (including hydrodynamic mass)-+ Differential Pressure.

The AISC 9th edition is used to qualify all of the components of the McGuireECCS Sump strainer except for stainless steel studs/bolts, which are qualifiedper ASME Section III, Division 1, NF-3324.6. Welds for stainless steel materialwere qualified per AWS D1.6.

3(k)(2) Summarize the structural qualification results and design marginsfor the various components of the sump strainer structuralassembly.

McGuire Response:Results of the structural analysis concluded that the design of the McGuiremodified ECCS sump strainer, including the Top-hats, the sections Inside theCrane Wall and in the Pipechase (including the connector piping and supports),the Inside the Crane Wall enclosure, and the Pipechase Vortex SuppressionRack, meets all AISC, AWS, and ASME code allowable stresses.

Page 65 of 86



3(k)(3) Summarize the evaluations performed for dynamic effects suchas pipe whip, jet impingement, and missile impacts associatedwith high-energy line breaks (as applicable).

McGuire Response:The response to RAI #31 of Enclosure 1 addressed the evaluation of dynamiceffects on the McGuire modified ECCS sump strainer. The response is repeatedhere for the convenience of the reader.

Postulated high energy pipe breaks which could potentially interact with themodified ECCS containment sump strainer were evaluated in accordance withMcGuire's current licensing basis as follows:

a. High energy Pipe Rupture composite drawings were reviewed to identify anypostulated breaks in close proximity to the modified ECCS sump strainerassembly and associated enclosure. Postulated high energy break locations,for the purpose of selecting rupture restraints, incorporated Leak BeforeBreak criteria.

b. Postulated breaks were evaluated to determine if the modified ECCScontainment sump strainer and associated enclosure were within the targetzone of pipe whip or jet impingement.

Per the above methodology, one interaction per Unit at McGuire was identifiedbased upon the new locations of the modified strainer assemblies.

Regulatory commitments were made to install pipe rupture restraints on theResidual Heat Removal System of each Unit. The required rupture restraint hasbeen installed on McGuire Unit 1; the required rupture restraint will be installedon Unit 2 in spring 2008.

3(k)(4) If a backflushing strategy is credited, provide a summarystatement regarding the sump strainer structural analysisconsidering reverse flow.

McGuire Response:As identified in the response to RAI #33 of Enclosure 1, the use of backflushing(or other active mitigative strategies) was not considered feasible for the McGuiremodified ECCS strainer design.

Page 66 of 86



3(l) Upstream Effects

The objective of the upstream effects assessment is to evaluate theflowpaths upstream of the containment sump for holdup of inventory whichcould reduce flow to and possibly starve the sump.

Provide a summary of the upstream effects evaluation including theinformation requested in GL 2004-02 Requested Information Item 2(d)(iv).

GL 2004-02 Requested Information Item 2(d)(iv)The basis for concluding that the water inventory required to ensure adequateECCS or CSS recirculation would not be held up or diverted by debris blockageat choke-points in containment recirculation sump return flowpaths.

3(l)(1) Summarize the evaluation of the flow paths from the postulated.break locations and containment spray washdown to identifypotential choke points in the flow field upstream of the sump.

McGuire Response:The evaluation of post-accident ECCS sump inventory holdup in the McGuirecontainments includes physical diversions (e.g., curbs and filled CS piping) aswell as potential debris blockage. The minimum ECCS sump pool level toensure strainer submergence is discussed in detail in the response to RAI #41 ofEnclosure 1, including the assumptions for lost inventory due to physicaldiversions. The potential loss of ECCS sump inventory due to debris blockage isdiscussed here.

The lower containment at McGuire is basically made up of two compartments -the area inside the Crane Wall and the Pipechase. These two areas areconnected at lower elevations by a number of crane wall penetrations on eachUnit, ranging in diameter up to 12 inches. Many of these penetrations are abovethe floor. Although it is possible for some of these penetrations to clog withdebris, it is unlikely that a sufficient number of the penetrations would becomeclogged sufficiently to create a situation where the ECCS sump could be starved.The computational fluid dynamics (CFD) model used for the evaluation of debristransport (discussed in detail in Section 3(e) of Enclosure 2) provides the basisfor this engineering judgment.

Other potential choke points include the ice condenser drain lines and therefueling canal drains. McGuire has a total of twenty 12-inch ice condenser drainlines for draining the melting ice. If one of these drains were to become clogged,the water would flow to the other drains. It is not likely that all 20 drains wouldbecome sufficiently clogged with debris to keep the water from flowing to the

Page 67 of 86



containment sump pool. The refueling canal in each Unit has six 8-inch drainsthat are open during operation. Four of the drains discharge inside the cranewall, and the other two discharge into the pipe chase. The plant was designedso that the majority of the upper containment spray water flows to lowercontainment through these six drains. Given the size of these drains and thedebris postulated to be washed down with the sprays (latent debris, paint chipsand/or particulate, and possibly a small quantity of LOCA generated fines blownpast the ice baskets) these drains are not likely to become clogged. Finally, theMcGuire debris generation calculation does not postulate significant amounts ofdebris being generated in upper containment, since this area is outside thelimiting break zone of influence.

3(l)(2) Summarize measures taken to mitigate potential choke points.

McGuire Response:McGuire Technical Specification Surveillance Requirement (SR) 3.5.2.8 requiresthat the ECCS sump be visually inspected to verify there are no restrictions as aresult of debris, and no evidence of structural distress or abnormal corrosionpresent prior to declaring the ECCS sump operable. A visual inspection ofcontainment is performed to ensure no loose material is present which could betransported to the Containment Sump and cause restriction of the ECCS pumpsuction during accident conditions prior to the transition from Mode 5 to Mode 4operations. When these inspections are performed, major outage work iscomplete, and any remaining loose material in containment must be logged andtracked in accordance with station procedures for control and accountability. Ifany debris, damage or deficiency were to be discovered during the inspection,station processes require entry into the corrective action program, with therequisite investigation and implementation of appropriate corrective action priorto the transition from Mode 5 to Mode 4.

McGuire Technical Specification 3.6.15 applies to the ice condenser drains andthe refueling canal drains. An inspection of the refueling canal drain is requiredto ensure that each canal drain valve is locked open and each drain is notobstructed by debris prior to entering Mode 4 from Mode 5 after partial/completefill of the canal. A visual inspection is performed every 92 days to verify that nodebris is present in the upper compartment or refueling canal that could obstructthe refueling canal drains. Lastly, each ice condenser floor drain valve is visuallyinspected and physically tested every 18 months to ensure it is not impaired byice, frost or debris, the valve seat shows no evidence of damage, the valveopening force is not excessive, and the drain from the ice condenser floor to thelower compartment is unrestricted.

Page 68 of 86



An additional mitigative measure was taken on the McGuire refueling canalbottom drains to reduce the potential for a choke point at these locations. Aperforated plate that existed on the bottom drain in the deep end of the refuelingcanal was removed for both Units 1 and 2. These modifications are alsoidentified in Section 3(j), item 3(j)(2) of Enclosure 2.

3(l)(3) Summarize the evaluation of water holdup at installed curbsand/or debris interceptors.

McGuire Response:The evaluation of post-accident EGOS sump inventory holdup in the McGuirecontainments includes physical diversions (e.g., curbs and filled CS piping) aswell as potential debris blockage. The minimum EGOS sump pool level toensure strainer submergence is discussed in detail in the response to RAI #41 ofEnclosure 1, including the assumptions for lost inventory due to physicaldiversions.

3(l)(4) Describe how potential blockage of reactor cavity and refuelingcavity drains has been evaluated, including likelihood of blockageand amount of expected holdup.

McGuire Response:

See responses to items.3(I)(1), 3(l)(2), and 3(l)(3) above.

Page 69 of 86



3(m) Downstream effects - Components and SystemsThe objective of the downstream effects, components and systems sectionis to evaluate the effects of debris carried downstream of the containmentsump screen on the function of the ECCS and CSS in terms of potentialwear of components and blockage of flow streams. Provide the informationrequested in GL 04-02 Requested Information Item 2(d)(v) and 2(d)(vi)regarding blockage, plugging, and wear at restrictions and close tolerancelocations in the ECCS and CSS downstream of the sump.

GL 2004-02 Requested Information Item 2(d)(vy)The basis for concluding that inadequate core or containment coolingwould not result due to debris blockage at flow restrictions in the ECCSand CSS flowpaths downstream of the sump screen, (e.g., a HPSI throttlevalve, pump bearings and seals, fuel assembly inlet debris screen, orcontainment spray nozzles). The discussion should consider the adequacyof the sump screen's mesh spacing and state the basis for concluding thatadverse gaps or breaches are not present on the screen surface.

GL 2004-02 Requested Information Item 2(d)(vi)Verification that the close-tolerance subcomponents in pumps, valves andother ECCS and CSS components are not susceptible to plugging orexcessive wear due to extended post-accident operation with debris-ladenfluids.

General Response Note:

On December 20, 2007, the NRC issued a Safety Evaluation for WCAP-16406-P,Revision I "Evaluation of Downstream Sump Debris Effects in Support of GS-191 ", dated August 2007. Duke previously evaluated the downstream effects ofsump debris on McGuire components and systems (as defined above) inaccordance with WCAP-16406-P, Revision 0, dated June 2005. A comparativeevaluation will be performed to address any differences extended by WCAP-16406-P, Revision I and the conclusions submitted to NRC by April 30, 2008 perNRC letter to Duke dated December 28, 2007. The responses and conclusionsthat follow, based on the original WCAP-16406-P, Revision 0 evaluation, areconsidered conservative and are not expected to change significantly sincefollow-on plant-specific testing modified the McGuire ECCS strainer design tofurther reduce the effect of downstream debris on components, systems, andfuel.

Page 70 of 86



3(m)(1) If NRC-approved methods were used (e.g., WCAP-16406-P withaccompanying NRC SE), briefly summarize the application of themethods. Indicate where the approved methods were not used orexceptions were taken, and summarize the evaluation of thoseareas.

McGuire Response:An evaluation of the downstream effects (i.e., ECCS sump strainer debrisbypass) of post-accident containment sump pool debris on the McGuireECCS/CS systems was performed by Westinghouse. The evaluation consideredthe effect of debris ingested through the containment sump strainer on thefollowing components that are required to operate:

* ECCS and CSS Valves" ECCS and CS Pumps* RHR and CSS Heat Exchangers" ECCS and CSS orifices" CSS spray nozzles and RHR auxiliary spray nozzles* Piping and instrumentation tubing

The evaluations, which included the Charging, Safety Injection, Residual HeatRemoval, and Containment Spray Systems, are based on the methodologydeveloped and documented in WCAP-16406-P, Revision 0, and considers thepotential effect on the aforementioned components of erosion, abrasion, and thepotential blockage of flow paths.

General Methodology Application Assumptions* The McGuire ECCS sump strainer top-hat module hole size is 3/32 inches

(0.09375 inches). Thus, the debris size for hard objects is determined to be0.09375 inches, based on the methodology outlined in Section 5.5 andAppendix J of WCAP-16406-P, Revision 0. Deformable objects of up to twotimes the strainer hole size by 1.1 times the strainer hole size are assumedto pass through the strainer, and are assumed to deform to pass throughany downstream clearance equal to or larger than the sump strainer holesize.

" For McGuire, the ECCS mission time for ECCS/CS components is assumedto be 30 days or 720 hours, as described in Section 8 of WCAP-1 6406-P,Revision 0.

The failure modes included in the pump evaluation are only those related tothe pump itself (i.e., they do not include the motor, gearboxes, couplings,etc.), since the debris loading in the pumped fluid is assumed to only affectthe internal components of the pump that are in contact with the pumpedfluid.

Page 71 of 86

Enclosure 2Information Addressing Issues Identified in Staff Content Guide for Ge'neric Letter


Debris Size Assumptions* Fibrous debris and RMI particulate debris are assumed to deplete per the

adjusted wear model presented in the Addenda to Appendix F of WCAP-16406-P, Revision 0.

" All other particulate debris and the failed coatings debris within the breakZOI are assumed to be less than 100 microns due to the characteristic sizespresented in the NRC SER for NEI 04-07, and as such will not deplete.

* Unqualified coatings in containment are assumed to fail in a size distributionwith 94% of the unqualified coatings debris greater than 400 microns whichwill therefore deplete; 4.5% are less than 400 microns but greater than 100microns, and 1.5% are less than or equal to 100 microns, and these smallerparticulates will not deplete. The size of the unqualified coatings debris lessthan 100 microns is assumed to be 50 microns on average.

Erosive and Abrasive Wear Model Assumptions" When applying the wear caused by the debris ingested through the ECCS

sump strainer, design conditions are assumed for the equipment with theexception of the pumps, where normal wear is taken into account.

" Per WCAP-1 6406-P, Revision 0 methodology, the abrasive and erosivewear on pumps used for service during normal plant operation is assumedto not exceed 3 mils.

* Per WCAP-1 6406-P, Revision 0 methodology, a debris depletion factor (A)of 0.07 hr-1 is assumed for both abrasive and erosive wear, which accountsfor the depletion of the sump pool debris.

" For the evaluation of wear on pumps, debris particles 50 microns andsmaller are assumed to cause only erosive wear on the pump internals. Thedesign running clearances in the ECCS/CS pumps typically range from0.010 to 0.023 inches. The smallest clearance in these pumps is the radialgap, which is 0.005 inches (5 mils). Debris particles smaller than 50microns are approximately 40% of this radial clearance and are thereforeunlikely to cause abrasive wear.

* Debris particles greater than 50 microns are conservatively assumed tocause abrasive wear of the pump internals.

Methodology ExceptionsThe SER for NEI 04-07 contains a requirement for licensees to assume that allcoatings in containment fail as 10 micron diameter spherical particulates.Although this requirement is conservative when evaluating head loss across theECCS sump strainer for which a "thin bed" effect is possible, it is notconservative when evaluating wear on components and valves.

Page 72 of 86



The Westinghouse wear evaluation of ECCS valves and components assumesan unqualified coating particulate size distribution that varies from 110% of theECCS sump strainer (top-hat) opening to 10 microns. This assumption isreasonable and conservative when evaluating the impact of unqualified coatingsparticulate on component and valve wear. There is significant public domaindocumentation that shows that coatings outside the conditions defined in thebreak ZOI will tend to fail at sizes larger than their constituent pigment size.

3(m)(2) Provide a summary and conclusions of downstream evaluations.

McGuire Response:The following conclusions and recommendations result from the downstreamevaluation of the McGuire ECCS/CS components:

ECCS/CS ValvesHaving a strainer hole size of less than 0.1 inch is.needed to avoid the potentialfor valve plugging. Of the McGuire valves that were identified as being critical tooperation following a LOCA in which the ECCS recirculation mode would berequired, none have the potential for plugging with the installed McGuire ECCSstrainer top-hat module hole size of 0.09375 inch.

The ECCS/CS valves identified as being of potential concern for sedimentationwere evaluated, and a calculation of the flow velocity through these valvesdetermined that sedimentation is not a concern.

The ECCS/CS valves evaluated for erosive wear (i.e., throttled valves) weredetermined to have flow area erosion increases that remain below theacceptance criteria, and therefore erosive wear is not a concern.

ECCS/CS PumpsThe pump hydraulic wear evaluation shows that none of the McGuire ECCS/CSpump wear gaps increase to the point of causing a hydraulic performanceconcern.

The pump mechanical evaluation for the Safety Injection and Charging multi-stage design pumps determined that the wear ring gap increases, due to theaction of erosive and abrasive debris, are less than the available wear margin of15 mils. As a result, there is no expected pump vibration concern.

The mechanical shaft seal assembly performance evaluation determined that thecarbon/graphite backup seal bushings for the ECCS/CS pumps are vulnerable ifexposed to the debris-laden sump pool fluid. The backup seal bushings are onlyrequired if failure of the primary pump seal is a concern. The primary pump sealis evaluated as unlikely to fail within the ECCS mission time, and since McGuiredose analyses credit the Engineered Safety Feature atmospheric filtration system

Page 73 of 86



located in the Auxiliary Building, there is no requirement to consider a pump sealfailure. Thus, no change to the pumps is required.

ECCS/CS Heat ExchangersThe McGuire CSS/RHR heat exchanger tube plugging evaluation demonstratedthat the tube inner diameter is larger than the anticipated debris particle size.Consequently, tube plugging will not occur. The heat exchanger wear evaluationdemonstrated that, because the actual wall thickness minus the thickness lost toerosion is greater than the wall thickness required to retain system pressure,tube failure due to erosion will not occur per the discussion in Section 8.3 ofWCAP-16406-P, Revision 0.

ECCS/CS Nozzles and OrificesThe McGuire CSS/RHR auxiliary spray nozzle plugging evaluation demonstratedthat the bore diameter is larger than the anticipated debris particle size.Consequently, plugging will not occur. For the spray nozzle wear evaluation, theincrease in spray nozzle flow rate due to an increased orifice diameter remainsbelow the acceptance limit specified in WCAP-16406-P, Revision 0, so nozzlewear is not a concern.

The ECCS/CS orifice plugging evaluation demonstrated that no orifice bore sizeis smaller than the largest particle that could pass through the McGuire ECCSstrainer top-hat module hole size of 0.09375 inch.

The ECCS/CS orifice wear evaluation identified that the worst case (i.e., the SIpump Cold Leg injection orifice) results in flow increasing by only 2.6%, whichremains below the acceptance criterion specified in WCAP-16406-P, Revision 0.Therefore, erosive wear of ECCS/CS orifices is not a concern.

ECCS/CS Instrument LinesThe McGuire ECCS/CS instrumentation tubing evaluation demonstrated that thetransverse recirculation flow velocities for instrumentation locations in the ECCSand the CSS are greater than 2.94 feet per second, which is above theacceptance criterion specified in WCAP-16406-P, Revision 0. Consequently,failure of the ECCS/CS instrumentation due to debris settlement does not occur.

3(m)(3) Provide a summary of design or operational changes made as aresult of downstream evaluations.

McGuire Response:The results of the McGuire downstream debris effects evaluations on the criticalECCS/CS components demonstrate that the currently installed components areacceptable for the expected ECCS mission time. No design or operationalchanges are required.

Page 74 of 86



3(n) Downstream Effects - Fuel and Vessel

The objective of the downstream effects, fuel and vessel section is toevaluate the effects that debris carried downstream of the containmentsump screen and into the reactor vessel has on core cooling.

General Response Note:

On November 21, 2007, the NRC issued a revision to the "Content Guide forGeneric Letter 2004-02 Supplemental Responses"' wherein the reference toWCAP-16793 in Section 3(n) was footnoted. The footnote indicated that staffevaluation guidance (in the form of a draft SER) was expected to be available toLicensees in December 2007. As this draft guidance has not yet been submitted,Duke will address the issues in this Section based on the in-vessel debrisevaluations performed, with a comparison to the original WCAP-16793-NP, Rev.0 methodology.

3(n)(1) Show that the in-vessel effects evaluation is consistent with, orbounded by, the industry generic guidance (WCAP-16793), asmodified by NRC staff comments on that document. Brieflysummarize the application of the methods. Indicate where theWCAP methods were not used or exceptions were taken, andsummarize the evaluation of those areas.

McGuire Response:As identified in Section 3(m), an evaluation of the downstream effects (i.e., ECCSsump strainer debris bypass) of post-accident containment sump pool debris onthe McGuire ECCS/CS systems was performed by Westinghouse. Theevaluation considered the effect of debris ingested through the containmentsump strainer on ECCS/CS components that are required to operate in theECCS recirculation mode.

The evaluations, which are based on the methodology developed anddocumented in WCAP-1 6406-P, Revision 0, also consider the potential effects ofdownstream debris-laden sump pool fluid on the flow paths through the reactorvessel internals and the nuclear fuel.

A summary of the assumptions used in the application of the WCAP-16406-P,Revision 0 methodology for the evaluation, and the exceptions taken to thismethodology, are located in Section 3(m) of Enclosure 2.

The results of the Westinghouse evaluation of the McGuire reactor vesselinternals, described following, reflect the methodology described in WCAP-16406-P, Revision 0.

Page 75 of 86



Reactor Vessel InternalsThe smallest flow clearance found in the McGuire reactor vessel internalsevaluation is 1.50 inches. The installed McGuire ECCS sump strainer top-hatmodules, with 0.09375 inch holes, thus will prevent plugging by either deformableor non-deformable debris.

Additionally, low flows in the lower reactor vessel plenum, combined with the factthat the Reactor Vessel Level Indication System (RVLIS) impulse lines are dead-ended, will prevent both the entry of debris into the RVLIS connection and thecollection of debris that might affect the differential pressure transmitters, perWCAP-16406-P, Revision 0 methodology. Therefore, debris ingested throughthe ECCS sump strainer and settling in the lower plenum of the reactor vesselwill not affect RVLIS water level measurements.

A separate plant-specific debris bypass evaluation of the McGuire modifiedECCS sump strainer design including the top-hat modules (with the DebrisEliminator feature), was performed to determine the size and quantity of fiberdebris that might bypass the strainer and enter the nuclear fuel assemblies.

Nuclear Fuel Assemblies

Westinghouse preliminarily evaluated the quantity of fiber that might reach thenuclear fuel assemblies during containment sump recirculation. According to thisevaluation, if the fiber size and quantity reaching the top or bottom of theMcGuire nuclear reactor core is sufficient to develop a fiber bed with a thicknessof 1/8-inch, this thin fibrous debris bed could filter out particulate debris thatbypasses through the ECCS containment sump strainer and result in a debrisbed with very low porosity. The low porosity through the debris bed wouldreduce or potentially block the flow passing through the fuel assemblies (i.e., thethin bed effect). This phenomenon is also discussed in the NRC SER for NEI 04-07. A fiber bed of 1/8-inch is utilized in the evaluation because thinner fiber bedswill not provide the required structure to bridge over the passageways at thebottom and top of the nuclear fuel assemblies.

McGuire's ECCS strainer top-hat design includes a Debris Bypass Eliminatorfeature, designed to reduce both the fibrous debris size and quantity that couldpotentially enter the core downstream of the sump strainer. The effectiveness ofthe Debris Bypass Eliminator feature was tested, using Nukon® fiber insulation,at various flowrates and fiber/particulate debris bed mixtures consistent with theappropriate McGuire debris transport evaluation.

ECCS Sump Strainer Fiber Bypass Evaluation MethodThe potential exists for small gaps/openings between the top-hat modules andplenums as well as other locations, which might allow bypass of fibrous material.Before determining the quantity of fiber that will pass through the gaps/openings

Page 76 of 86



within the McGuire sump strainer plenums, the surface area of thesegaps/openings is determined. Using this surface area and the test data fromNRC-sponsored bypass testing (as documented in the Los Alamos ScreenPenetration Test Report LA-UR-04-5416), the quantity of fiber passing throughthe gaps/openings is determined.

Additionally, the particulate/fibrous debris bypass through a strainer top-hatmodule equipped with the Debris Bypass Eliminator was independently assessedand measured in a flume test, including a resulting debris characteristicevaluation.

Using both the tested fibrous debris bypass through the gaps/openings within thesump strainer plenums, and the tested debris bypass through the strainer top-hatmodules, the potential for blockage of the nuclear fuel assemblies duringcontainment sump recirculation is evaluated.

Bypass Evaluation Assumptions" It is conservatively assumed that all of the gaps between the cover plates and

the plenums as well as the clearance between the top-hats and the plenumsare equal to the 1/16-inch clearance specified on the design drawings. Theclearance specified is the maximum allowable clearance. Most of theconnected components will have little or no clearance between them. Atmany of the connections, two right-angle flow direction changes are requiredto allow a fiber to pass through a potential gap, thus increasing the likelihoodof the fiber being trapped within the gap.

* The Nukon® and Thermal-Wrap® fiber insulation installed on the piping withinthe McGuire containment has a density of 2.4 lb/ft3. It is conservativelyassumed that the density of the fiber when reaching the top or bottom of thereactor core is unchanged. Typically, Nukon® and Thermal-Wrap® fibrousinsulation compresses to a greater density when it builds a debris bed mixedwith particulates.

* It is assumed that the wire ropes used in sealing between plenums installedinside the crane wall eliminate all bypasses through these gaps. This isreasonable because the sealing plates will be fastened tightly to the plenumswith the wire ropes acting as gaskets. The plenums installed outside thecrane wall are sealed with metal bands that are tightened, and thencompressed by additional plates on the sides. This creates a tight seal withvirtually no gap for fiber to pass through.

* Further assumptions regarding debris transport are detailed in the responseto RAI #1 of Enclosure 1, and in Section 3(e) of Enclosure 2.

Page 77 of 86



Bypass Evaluation Results SummaryThe quantity and length of the fibers passing through the McGuire top-hat'strainer modules with the Debris Bypass Eliminator feature was measured.Based on these measurements, the majority of the fiber bypass (over 98%)through the debris bypass eliminator will not build a fiber bed below or above thenuclear reactor core due to the short length of the fibers.

The quantity of fiber passing through various 1/16-inch gaps and 3/32-inchopenings within the modified ECCS sump strainer design for McGuire Units 1and 2 was conservatively determined. The total quantity of fiber that couldbypass through these gaps/openings is not sufficient to develop a thin bed ofdebris with a thickness of 1/8-inch at the top or bottom of the nuclear reactorcore. The total amount of fiber bypassed cannot provide the required structure tobridge over the passageways at the bottom and top of the nuclear fuelassemblies. Therefore, per this evaluation, sufficient open flow paths will exist forcooling of the nuclear fuel assemblies.

WCAP-1 6793-NP, "Evaluation of Long-Term Cooling Considering Particulate andChemical Debris in the Recirculating Fluid," Revision 0, dated May 2007,provides analyses for assessing the effects of fibrous, particulate, and chemicaldebris on nuclear fuel assemblies.

The existing guidance provided to the industry is used in WCAP-16793-NP toprovide the framework for the analyses that further address these concerns.Existing guidance incorporated in WCAP-16793-NP:

* WCAP-16406-P: Section 9.0 (cold leg injection, hot leg injection, fiber,particulates, etc.), including Addenda

* NEI.04-07, Volume 1: Section 7.3

" NEI 04-07, Volume 2: Section 7.3

* Draft NRC Staff Review Guidance for "Evaluation of Downstream Effects ofDebris Ingress into the PWR RCS on Long Term Core Cooling Following aLOCA", dated November 22, 2005.

While the guidance in these documents provides information regarding how toassess the effects of debris on fuel, the application of these methods is stated inWCAP-16793-NP as being significantly conservative. In particular, the followingconservative assumptions in the existing guidance are identified:

• All debris that penetrates the sump strainer reaches the core. Further, allfibrous debris is neutrally buoyant and is long enough to be captured at thecore inlet.

Page 78 of 86



* The formation of a thin bed (i.e., a bed of fiber 1/8-inch thick) at the core inletis sufficient to preclude flow to the core based on the results presented inNUREG/CR-6224.

The WCAP then describes various evaluations and tests performed to determinethe likelihood that bypass debris will adversely affect nuclear fuel assemblies,including a debris characteristics evaluation of industry-representative fibers andparticulates.

Summarizing these tests and evaluations, reasonable assurance of long-termcore cooling for all plants is demonstrated in WCAP-16793-NP by the following:

* The size of holes in replacement sump strainer designs limits the size ofdebris that is passed through the strainer during operation of the ECCS inthe recirculation mode.

* Based on test observations, the characteristic dimension of this debris istypically less than the strainer hole size, even for fibrous debris.Consequently, debris buildup at critical locations in the reactor vessel andcore is not expected.

" Based on data presented internationally during the resolution of the BWRstrainer performance concerns, fibrous debris was observed to not stronglyadhere to fuel cladding. Thus, the small size of the debris and its tendencyto not adhere to fuel indicates that long-term core cooling of the fuel will notbe impaired by either the collection of fibrous and particulate debris in fuelelements, or by the collection of fibrous debris on fuel cladding surfaces.

• Supporting calculations have demonstrated long-term core cooling will bemaintained with about 99.4% of the core blocked. The cladding temperatureresponse to blockage at grids and the collection of precipitation on cladsurfaces was also demonstrated to be acceptable with resulting claddingtemperatures less than 400'F.

The McGuire plant-specific nuclear fuel assembly debris evaluation is performedvia testing and analyses that incorporate the previously available industryguidance and conservatisms. Using these techniques and acceptance criteriathe McGuire modified ECCS sump strainer is shown to be capable of preventingan adverse build-up of fibrous debris on the core. Additionally, the debrischaracteristics evaluation of the fibers deposited downstream of the McGuireECCS sump strainer indicates that the downstream WCAP-16793-NP debrischaracteristics are more limiting in size.

As such, the plant-specific fibrous debris bypass evaluations performed forMcGuire are bounded by the evaluations described by WCAP-1 6793-NP.

Page 79 of 86



Duke is aware that NRC is still evaluating the industry guidance provided byWCAP-16793-NP, and will monitor the status of this evaluation. Based on theresults of the Duke-specific downstream fiber and particulate debris effectsevaluations performed, significant changes to the preceding assessment of theMcGuire ECCS sump strainer are not expected.

The assessment of downstream chemical effects on the nuclear fuel assembliesis also described by WCAP-1 6793-NP analysis methodology. This issue isaddressed for McGuire in Section 3(o) of Enclosure 2.

Page 80 of 86

)



3(o) Chemical Effects

The objective of the chemical effects section is to evaluate the effect thatchemical precipitates have on head loss and core cooling.

General Response Note:On November 21, 2007, the NRC issued a revision to the "Content Guide forGeneric Letter 2004-02 Supplemental Responses"' wherein the reference to bothWCAP-16530-P and WCAP-16793 were footnoted. The footnotes indicated thatstaff evaluation guidance for these two documents (in the form of draft SERs)was expected to be available to Licensees in November and December 2007. Atthis time, only the draft SER for WCAP-16530-P has been issued. In it, NRCidentified that WCAP-16793-NP, which specifically addresses the chemicaleffects concerns in Section 3(o), was still under review.

Duke will address the long-term core cooling issues identified in this Sectionbased on the in-vessel chemical effects evaluation performed using the WCAP-16793-NP methodology issued in May 2007.

3(o)(1) Provide a summary of evaluation results that show that chemicalprecipitates formed in the post-LOCA containment environment,either by themselves or combined with debris, do not deposit atthe sump screen to the extent that an unacceptable head lossresults, or deposit downstream of the sump screen to the extentthat long-term core cooling is unacceptably impeded.

McGuire Response:Chemical Deposition at the ECCS Sump StrainerBenchtop and Vertical Loop Test chemical effects testing performed by Duke isdiscussed in detail in the response to RAI #10 of Enclosure 1. Further chemicaleffects testing via the Integrated Prototype Test (IPT), designed to quantify theconsequence of chemicals in the McGuire containment sump pool on the ECCSsump strainer debris bed head loss, is discussed in detail in the response to RAI#11 of Enclosure 1. The results of the IPT are being finalized and will besubmitted with the supplemental response to RAI #12 of Enclosure 1. This is acommitment identified in Duke letter dated November 6, 2007, "Request forExtension of Completion Dates for McGuire Units 1 and 2 Corrective ActionsRequired by NRC Generic Letter (GL) 2004-02", and the associated NRC SERdated December 28, 2007. Per this commitment Duke expects to have thisdocumentation, and to supplement the response to RAI #12, by April 30, 2008.

Page 81 of 86



Chemical Deposition Downstream of the ECCS Sump StrainerThe chemical reactions of most concern for core deposition are those thatrelease material into solution in a form where it can bypass the ECCS sumpstrainer, collect in the reactor vessel, and precipitate on heated fuel claddingsurfaces. The chemical reactions leading to the generation of such transportablematerial follow:

" Corrosion or dissolution of system materials to directly produce a hydrouscorrosion product that does not settle.

* Corrosion or dissolution of system materials to produce dissolved materialthat later forms precipitates on the fuel due to temperature change and/or pHchange.

" Corrosion or dissolution of system materials followed by chemical reactionswith other coolant chemicals to produce hydrous precipitates that do notsettle.

Corrosion or dissolution of system materials is a first step that is common to all ofthe reactions. The assessment of precipitation or deposition reactions within thepost-LOCA environment must be able to estimate the dissolution behavior ofcontainment materials.

Westinghouse previously developed a method for predicting post-LOCA chemicalreactions and the formation of material that could affect ECCS sump strainers inWCAP-16530-NP. This methodology has been reviewed by the NRC, andMcGuire utilized it as a basis for demonstrating adequate ECCS sump strainerperformance in the Integrated Prototype Test (IPT) for chemical effects describedin the responses to RAI #10 and RAI #11 of Enclosure 1.

Recent NRC concerns related to post-LOCA chemical reactions have focused onthe core. Specifically, the NRC identified that they expected the followingchemical effects concerns be addressed:

* Assessment of chemical concentration effects due to long-term boiling

" Consideration of plate-out of deposits on the fuel rods

* Estimated effect of deposits on core heat transfer

The LOCA Deposition Analysis Model (LOCADM), described in WCAP-16793-NP, was developed to enable all plants, regardless of NSSS vendor(Westinghouse, CE or B&W) to address these concerns when documenting theviability of long-term core cooling.

Page 82 of 86



WCAP- 16793-NP AssumptionsThe deposition method makes several assumptions that are conservative and, asa result, the predictions of deposit thickness and fuel surface temperature areconsidered to be bounding rather than best-estimate.

1. Once formed, deposits will not be thinned by flow attrition or by dissolution.

2. No deposition takes place apart from the fuel heat transfer surfaces. A best-estimate approach would have accounted for deposition on non-fuelsurfaces such as the RHR heat exchangers and surfaces in containment,resulting in thinner core deposits.

3. The mass balance approach for determining material transport around theECCS does not take into account any moisture carryover in the steamexiting the reactor vessel. Experimental measurements simulating the post-LOCA environment indicate that concentration of non-volatile material withinthe reactor vessel will be considerably reduced if moisture carryover isincluded in the estimation. Not including boron and coolant impurities in themoisture carryover is conservative.

4. The effect of boiling point elevation due to the concentration of solutes is notcurrently modeled. This simplification will result in an over-prediction ofboiling in the core and thus any error introduced by the simplification will bein the conservative direction.

5. Only species that have dissolved into solution or species that have dissolvedand then precipitated into suspended particles are considered. The transportof large debris particles from containment and re-deposition of debris fromfuel failures have not been included. Larger debris will either settle or will bephysically retained by the ECCS sump strainer, the fuel assembly inletdebris filters, or in other locations where flow is restricted. This mode ofblockage is addressed in Section 3(f), Section 3(m), and Section 3(n) ofEnclosure 2.

6. All impurities transported into a deposit by boiling will be deposited at a ratethat is equal to the product of the steaming rate and the coolant impurityconcentration.

7. The non-boiling rate of deposit build-up is proportional to heat flux and is1/80th of that of boiling deposition at the same heat flux. This ratio is basedon empirical data for mixed calcium salts under boiling and non-boilingconditions.

8. The deposition of impurities on the fuel clad surface is assumed to bedistributed according to the core power distribution.

Page 83 of 86



WCAP- 16793-NP Evaluation ResultsEvaluation of chemical effects in the core region to form precipitation on thecladding surface was performed. Considering the variation in plant-specificchemistries, this evaluation was performed by extending the method of WCAP-16530-NP to estimate the potential for plate-out on the surface of fuel cladding.This method is available for all Westinghouse, CE and B&W plants to performplant-specific evaluations in which their plant-specific chemistry is accounted for.

Sample calculations were performed using particularly challenging plantchemistries, and fuel clad temperatures were predicted to remain below 400°Fover a 30-day period following the postulated event. Due to the interaction ofseveral of the parameters, WCAP-1 6793-NP suggests that plants perform aplant-specific evaluation by comparison to these sample calculations to confirmthat chemical plate-out on the fuel does not result in the prediction of fuelcladding temperatures approaching the 800°F acceptance basis value.

Comparison to the sample calculations presented in WCAP-16793-NP showsthat the predicted McGuire post-LOCA conditions and chemistry parameters arebounded by the WCAP analyses, and therefore the long-term cooling capabilityof the McGuire nuclear core is not impeded by downstream chemical effects.

WCAP-16793-NP and Boric Acid PrecipitationThe effect of sump debris and sump chemical compounds on boric acidprecipitation has been reviewed with respect to displaced liquid volume, thepotential impact on assumed mixing volumes, alternate flow paths, and chemicaleffects as it pertains to potential precipitates in the core. It is concluded thatsump debris and related chemical effects do not create a boric acid precipitationconcern and that the introduction of debris to the RCS does not significantlyaffect the current licensing basis boric acid precipitation calculations. Therefore,the current accepted licensing calculations that demonstrate appropriate boricacid dilution to preclude boric acid precipitation remain valid.

3(o)(2) Content guidance for chemical effects is provided in Enclosure 3to a letter from the NRC to NEI dated September 27, 2007 (ADAMSAccession No. ML0726007425).

McGuire Response:Enclosure 3 to the letter from NRC to NEI dated September 27, 2007("Evaluation Guidance for the Review of GSI-191 Plant-Specific Chemical EffectEvaluations"--ADAMS Accession No. ML072600372) is draft guidance for thestaff (and licensees) to ensure the chemical effects portions of Generic Letter2004-02 plant-specific evaluations appropriately address the chemical effectsthat can occur following a postulated loss of'coolant accident (LOCA). Thisguidance invokes industry testing methodology and observations of industry

Page 84 of 86



testing to facilitate the process of assessing potential concerns, formulating plansof testing, conducting tests, and evaluating test results.

As noted previously, Duke's strategy for addressing chemical effects on theMcGuire modified ECCS sump strainer are addressed in the responses to RAI#10 and RAI #11 of Enclosure 1, which describe the preliminary and plant-specific chemical effects testing (Integrated Prototype Test) performed and theindustry-related bases for the development of the tests.

Page 85 of 86



3(p) Licensing Basis

The objective of the licensing basis section is to provide informationregarding any changes to the plant licensing basis due to the sumpevaluation or plant modifications. Provide the information requested in GL04-02 Requested Information Item 2.(e) regarding changes to the plantlicensing basis. The effective date for changes to the licensing basisshould be specified. This date should correspond to that specified in the 10CFR 50.59 evaluation for the change to the licensing basis.

GL 2004-02 Requested Information Item 2(e)A general description of and planned schedule for any changes to the plantlicensing bases resulting from any analysis or plant modifications made toensure compliance with the regulatory requirements listed in theApplicable Regulatory Requirements section of this generic letter. Anylicensing actions or exemption requests needed to support changes to theplant licensing basis should be included.

McGuire Response:The discussion of the McGuire licensing basis requested in Section 3(p) isprovided in Section 1 of Enclosure 2, Overall Compliance.

Page 86 of 86

Documents

McGuire, Units 1 and 2, Response to Request for Additional ... · tags/labels) that transport to the ECCS sump are addressed in Enclosure 2, Section 3(d). The following assumptions